ERIK NORDEN^KTfOLD
^>^^
^
i^ u. \^^^
t
ID
Marine Biological laboratory Library
Woods Hole, Mass.
*^^«»'N*
Presented by
From the estate of
Louis W. Hutchins
M^, 1964
01
I
I
I
I
D
I
I
I
E3^^^^^^^^^^^^^E3E
THE HISTORY OF BIOLOGY
CHAPTERS IN
MODERN BIOLOGY
AND
BIOMETRICS
By Raymond Pearl :
THE BIOLOGY OF POPULATION GROWTH
ALCOHOL AND LONGEVITY
THE RATE OF LIVING
By Julian Huxley :
ESSAYS OF A BIOLOGIST
ESSAYS IN POPULAR SCIENCE
By William lAorton Wheeler :
FOIBLES OF INSECTS AND MEN
IKiillll' l,ili(l'!i'!<il, ';'lllll||'lilll!ljl!ir''(ll i l.'i'„|jlliri|l| iHllliIiMi llllliiliill'ril i 'W djlljl
N
I— I
Pi
<
o
<
<:
Pi
0
^
^
(A
CA^
^
U)
2
<
C/5
h-^
w
^
>
^
C/3
<
^
ft<
Q
2:
Q
<:
5
^
By ERIK NORDENSKIOLD
TRANSLATED FROM THE SWEDISH BY
LEONARD BUCKNALL EYRE
THE
HISTORY
OF
BIOLOGY
y /5-
A SURVEY
\"jLV MASS.
TUDOR PUBLISHING CO.
MCMXXXVi New York
COPYRIGHT 192.8 BY ALFRED A. KNOPF, INC.
ALL RIGHTS RESERVED NO PART OF THIS BOOK MAY BE REPRINTED
IN ANY FORM WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER
NEW EDITION AUGUST, I935
SECOND PRINTING JULY, I936
Originally issued as
BIOLOGINS HISTORIA
in three volumes
1920-24
STOCKHOLM, BJORCK & BORJESSON
MANUFACTURED IN THE UNITED STATES OF AMERICA
FOREWORD
This work, ivhlch is here presented in the English language, is
based on a course of lectures given at the University of Helsingfors,
Finland, during the academic year igi6-iy. It is the author s in-
tention to present a picture of the development of biological science
throughout the ages, viewed in conjunction ivith the general cultural
development of mankind. Regarded thus as a link in the general history
of culture, the problems of biology luill, it is hoped, prove of interest
not only to young university students, for whom this book is priinarily
intended, but also to a still wider public. With regard to modern times,
for obvious reasons it has only been possible in such a brief history as
this to give a very summary account of recent developments. A more
thorough knowledge of the results of specialised biological research will
be gained by reference to the literary works of professional biologists,
which often contain a historical survey by luay of introduction. On
the other hand, the theoretical principles on which research work has
been carried out have been discussed here in greater detail, both for the
reason that records of them are not so easily accessible and on account
of the influence they have exerted upon culture in general. In accordance
with this principle a number of typical representatives of each trend of
thought have been selected for inclusion and their ivork described, while
no attempt has been made to present a complete record of all personalities
that have figured in the biological world. In this, as in other historical
VI FOREWORD
works, the selection has of course been made by a process of elimination,
which to a certain extent was bound to be subjective; especially in a
work dealing mainly with a general historical development it has been
necessary to exclude the names of a great many brilliant specialists^ in
spite of the fact that their ivork may be of lasting value, while other
personalities, perhaps in themselves of less importance, have been men-
tioned on account of the part they have played in the getieral cultural
development of their period. For the same reason representatives of
scientific progress in the various civilised countries of the world have
been included, as far as space has alloived, in order to present as com-
prehensive an idea as possible of the progress of science and the contri-
butions that different peoples have made thereto.
For their assistance in preparing the English edition I take this
opportunity of recording my thanks to Mr. Leonard Bucknall Eyre,
B.A. Cantab., of Stockholm, who has translated the book from the
Swedish, and to Mr. Alfred A. Kfiopf, ivho has promoted its pub-
lication.
StockhoUn, Novetnber igzj The Author
CONTENTS
PART ONE
BIOLOGY IN CLASSICAL ANTIQUITY, THE MIDDLE
AGES, AND THE RENAISSANCE
I. The development of biology amongst the primitive peoples
and the civilized nations of the East 3
II. The earliest Greek natural philosophy 8
III. The earlier phase of Greek medical science and its significance
for the development of biology 15
IV. The end of natural-philosophical speculation. The prede-
cessors of Aristotle 30
V. Aristotle 34
VI. Natural-philosophical systems after Aristotle 45
VII. Specialized biological research after Aristotle 50
VIII. The decline of science in late antiquity 58
IX. Biological science among the Arabians 68
X. Biology during the Christian Middle Ages 74
THE HISTORY OF BIOLOGY DURING THE
RENAISSANCE
«;
XI. The end of mediaeval science 8z
XII. New cosmic ideas and new scientific method 84
XIII. Descriptive biological research during the Renaissance:
I. Zoography 91
X. Anatomy 98
8S070
Vlll CONTENTS
XIV. The discovery of the circulation of the blood:
I. Harvey's predecessors 108
i. Harvey 114
*
PART TWO
BIOLOGY IN THE SEVENTEENTH AND EIGHTEENTH
CENTURIES
I. The origin of the modern idea of nature in the seventeenth and
eighteenth centuries iii
II. The mechanical nature-systems 113
III. Mystical speculation upon natural science 131
IV. Biological research in the seventeenth century:
I. Harvey's successors 141
X. Attempts at a mechanical explanation of life-phenomena 151
3. Microscopies and microtechnology 158
V. Biological speculations and controversial questions at the
beginning of the eighteenth century 174
VI. The development of systematic classification before Linnasus 190
VII. Linnasus and his pupils 2.03
VIII. Buffon XI 9
IX. Invertebrate research in the eighteenth century 2.30
X. Experimental and speculative biology in the eighteenth
century X34
XI. Descriptive and comparative anatomy in the eighteenth
century X58
XII. The first beginnings of modern chemistry and its influence
upon the development of biology 2.64
XIII. Critical philosophy and Rc^mantic conceptions of nature:
I. Kant and his immediate successors 2.68
X. Goethe 2.79
CONTENTS IX
XIV. Natural philosophical biology:
I. Germany and Scandinavia 2.86
■3.. England and France 2.93
PART THREE
MODERN BIOLOGY
BIOLOGY DURING THE FIRST HALF OF THE
NINETEENTH CENTURY
I. From natural philosophy to modern biology:
I. The predecessors of comparative anatomy 301
2.. Humboldt 314
3. Lamarck 316
II. Cuvier 331
III. Bichat and his tissue theory 344
IV. Cuvier's younger contemporaries 35i
V. The progress of embryology 362.
VI. The development of experimental research and its application
to comparative biology 370
VII. Microscopy and cytology 389
VIII. The continued development of biology until the advent of
Darwinism:
I. Experimental research work 406
-L. Morphology and classification 414
3. Microbiology 42.6
4. Botany 435
IX. Positivist and materialist natural philosophy 441
*
FROM DARWIN TO OUR OWN DAY
X. The preconditions of Darwinism:
I . Modern geology 453
1. The ideal preconditions of Darwinism 458
X CONTENTS
XL Darwin ^gi
XII. For and against Darwin 477
XIII. The doctrine of descent based on morphological grounds.
Gegenbaur and his school ^og
XIV. Haeckel and monism 505
XV. Morphological specialized research under the influence of
Darwinism :
I. Anatomy and embryology 518
2- Cytology 533
3. Microbiology 544
4. Vegetable morphology 550
5. Geographical biology 558
XVI. Neo-Darwinism and neo-Lamarckism 562.
XVII. Experimental biology:
I. Experimental morphology 574
1. Experimental heredity-research 583
3. Biochemistry 594
4. Animal psychology 599
XVIII, Modern theoretical speculations:
I. Mechanism and vitalism 603
1. The idea of species and some problems in connexion
therewith 613
Sources and Literature
Sources 617
Literature 6x9
Index of Names follows page 6x9
LIST OF ILLUSTRATIONS
GIROLAMO FABRIZIO- | Frontispiece
ANDREAS VESALIUS •
ULISSE ALDROVANDI • 74
ANTONY VON LEEUWENHOEK • 74
GEORGE LOUIS LECLERC DE BUFFON • 108
WILLIAM HARVEY • 108
ALBRECHT VON HALLER • 131
PARACELSUS • 131
MARCELLO MALPIGHI • 190
GIOVANNI ALFONSO BORELLI • 190
NICOLAUS STENO • 134
JOHN RAY • 134
LINN^US • z68
CHARLES BONNET • 2.68
LORENZ OKEN • 2.86
LAMARCK • i86
CUVIER • 304
MARIE FRANCOIS XAVIER BICHAT • 304
ALEXANDER VON HUMBOLDT • 370
JOHANN WOLFGANG GOETHE • 370
KARL ERNST VON BAER • 414
JOHANNES PETER MULLER • 414
LIST OF ILLUSTRATIONS
CHARLES DARWIN • 458
CHARLES LYELL • 458
CARL GEGENBAUER • 498
RUDOLPH ALBERT VON KOLLIKER • 498
MENDEL • 561
RICHARD OWEN • 561
THOMAS HENRY HUXLEY • 594
OSCAR HERTWIG • 594
ERNST HEINRICH HAECKEL • 61I.
WILHELM LUDWIG JOHANNSEN • 6li
PART ONE
.BIOLOGY IN CLASSICAL ANTIQUITY,
THE MIDDLE AGES,
AND THE RENAISSANCE
CHAPTER I
THE DEVELOPMENT OF BIOLOGY AMONGST THE PRIMITIVE
PEOPLES AND THE CIVILIZED NATIONS OF THE EAST
Primitive man s speculations upon life
THE EARLIEST FOUNDATION of all our natural scientific knowledge is to
be sought in the observations of nature collected in the course of
thousands of years by prehistoric peoples who had reached a primitive
stage of civilization. This empirical folk-knowledge, which the student of
folk-lore in our own day investigates from a historical and national-psycho-
logical point of view, has not only been the starting-point for all scientific
thought, but has also, right up to the most recent times, to a certain extent
influenced scientific research itself; increased its store of facts with material
for observation and even now and then given rise to problems which science
has debated. Primitive man's speculations upon life have naturally been
influenced by his mode of life in various climates and under varying con-
ditions. Common to them all, however, would appear to have been the fact
that the first thing that has induced man to reflect upon life has been its
cessation: death. And to the aborigines what we call a natural death is
actually the most wonderful; that a man should fall in a fight against wild
animals or his enemies is all part of the order of the day, but that the powers
of a sound and healthy man should suddenly and without reason begin to
fail and life to cease with or without the accompaniment of pain — that is a
thing one finds it hard to acquiesce in. And the thing becomes all the more
remarkable for the fact that again and again at night the departed one ap-
pears in dreams to those who have survived him. These dreams have given
rise to a belief in ghosts, spectres, and spiritual powers of various kinds, both
friendly and evil, and this belief has in its turn called forth measures with a
view to deriving advantage from the well-disposed and avoiding the snares
of the wicked. Thus measures of many and various kinds were adopted in
regard to the bodies of the dead, which were either cremated or otherwise
destroyed in order to render it impossible for them to return amongst the liv-
ing, or else, on the other hand, they were elaborately cared for by the preser-
vation of the skeleton or by embalming, which was intended to make the
dead well-disposed towards their survivors. From these manipulations arose
the first knowledge of the anatomy of the human body, while observations
of the actual course of death created certain physiological ideas. Men learnt
3
4 THE HISTORY OF BIOLOGY
to observe the heart-beat and to connect life with its continuance or cessa-
tion, and thus the heart itself was regarded as the organ of life. Breathing
was also observed to be an essential condition of life, and in particular the
deep expiration which indeed so often attends the actual moment of death
gave rise to the idea of life as having something of the nature of air, being
dependent upon the respiratory organs and leaving the body through them.
In medieval church paintings this belief reappears in a particularly naive
manner: the soul of the dying is seen to leave the body in the form of a little
child creeping out through the mouth. Likewise the words of the biblical
story of the creation to the effect that God breathed into man's nostrils the
breath of life testifies to the same kind of idea. And so there arose, as a fur-
ther development of these ideas, the belief that the breath or spirit lives
when the body dies. The contrast between body and spirit which is an out-
come of the ideas described above is included in the speculations of the earliest
natural philosophers as a fundamental principle.
Kelationship to animals
However, in the mind of primitive man these lines of thought, proceeding
from the contrast between life and death, are crossed by others, which have
their origin in his relationship to the rest of the world of living creatures.
The great w'ld beasts — bears, lions, elephants — were difficult to overcome;
it was often necessary to try, as far as was possible, to make friends with
them. Other beasts were regarded with terror for their night-roving habits
and horrible cries, such as hyenas and owls; while some possessed otherwise
enviable natural gifts — the fox his cunning, the deer his swiftness, etc.
It is out of all this that we must explain the origin of the mass of animal
superstition that has filled the life of both wild and civilized peoples. As
forms in which this superstition has developed may be mentioned totem-
ism, or the custom existing among certain wild peoples of adopting animals
as a kind of guardian spirit and family symbol, as well as the belief in
and worship of holy animals, which, even amongst highly civilized peo-
ples, such as the Egyptians and the Romans, have played such an important
part in life. This animal superstition has naturally contributed towards
increasing the interest in and knowledge of animals, both as regards the
habits of life of those which were worshipped as gods, and the anatomy of
those which were offered in sacrifice and were most minutely examined
with a view to divining portents for the future from their internal structure.
Primitive surgery and medicine
Finally, a third extremely important source of biological knowledge has
been medical science. Primitive surgery, which originated in attempts to
cure various bodily injuries, must of course eventually lead to a certain
amount of knowledge of the anatomy of the human body, a knowledge
which was increased by the process of comparison with the experience
CLASSICAL ANTIQUITY, MIDDLE AGES 5
gained from the slaughter of wild and tame animals. As to the natural dis-
eases, the same holds good for these as what has just been mentioned in
regard to death; for lack of ability to explain them naturally, people took
refuge in a belief in supernatural causes. The belief in enchantments of various
kinds which arose therefrom and which has been maintained even amongst
civilized peoples for a surprisingly long time, fills one of the darkest chapters
in the history of civilization. Disasters of supernatural origin of course
demanded corresponding remedies, and consequently the earliest practice of
medical science among all races of mankind has been that of magic: they
sought to remove the evil by setting sorcery against sorcery. However, the
regular course of certain processes of disease could not fail to be observed
and conclusions drawn therefrom as to the functions of the body in sickness
and health. By a comparison of these observations a number of primitive ideas
were acquired on physiology and pathology. Hand in hand with this was
evolved the theory of pharmacology, based on experiments — originally for
the most part for magical purposes — with plants which experience proved
to be poisonous or otherwise capable of affecting the life-process. Through
observations of this kind the knowledge of life was still further enhanced.
It was not given, however, to just anyone to acquire all this knowledge, the
origins and development of which have been described above. The super-
natural and mysterious elements in them made them a privilege for cer-
tain qualified persons: magicians, sorcerers, sacrificial priests. Among these
classes of people they were shared and handed down as professional secrets,
until in course of time a division of them took place — the magical and ritual
customs became the professional sphere of the priests, while the amassed
knowledge of nature, released from the obstructive bonds of magic, was de-
veloped by independent inquirers into a free sphere of learning. The people
amongst whom this independent natural science first arose were the Greeks.
But long before Greek culture appears in history, the people of the East
had already bequeathed historical evidences of their civilization, and these
deserve all the more to be carefully examined for such contributions to bio-
logical knowledge as they may have to show, seeing that the whole of
Greek culture was so highly influenced by the oriental.
Babylonian science
The earliest home of human civilization is now generally supposed to have
been Babylon, and a high standard of culture was maintained there under
the dominion of various types of peoples up to the latter part of the Mid-
dle Ages. The "oriental wisdom" which has played such an important part
in the mystical literature of all times also originates from there through
a more or less varying number of intermediate stages. Actually, the mys-
tical and the magical have from the earliest times played a predominant
role in that country's learning, undoubtedly owing to the fact that all
6 THE HISTORY OF BIOLOGY
knowledge was nurtured and developed by a powerful priesthood. The
conception of nature was influenced thereby: the early knowledge of as-
tronomy was placed at the service of mystical powers, as were also mathe-
matics and medicine. The latter science, however, in certain respects made
no small progress. The knowledge of anatomy was considerable; preserved
clay-models of certain of the viscera of the body prove this and give evi-
dence that the dissection of corpses must have taken place in spite of the
horror which Orientals have always felt for the dead and their spirits.
It is clear from preserved writings on medicine that the heart was re-
garded as the organ of intelligence, and the liver as that of the blood-cir-
culation; the blood was divided into "light" and "dark" blood — arterial
and venous. The knowledge of higher animal forms was, as extant lists of
nomenclature go to prove, quite considerable, and kings and princes kept
rare live animals in their gardens. Even animal-doctors are mentioned in
preserved inscriptions.
Egyptian medicine and natural knowledge
Again, in the other oldest civilized country of the West, Egypt, there was
developed at an early period an art of healing which was based not merely
upon superstition, but also upon actual observations. The early perfected
religious practice of preserving dead bodies from putrefaction by con-
serving the skeleton and, later, by embalming offered an opportunity of
acquiring anatomical knowledge which proved of great benefit to medical
science. The sacred animals were likewise studied with minute care, and
writings have been discovered giving in detail the history of the develop-
ment of the sacred scarab, and even the metamorphosis of the frog and the
fly. The parasitic worms that so infected Egypt were also objects of investi-
gation and speculation.
Israelitic conception of nature
With regard, finally, to the Israelitic people, their cultural contribution
has been in a sphere entirely different from the natural-scientific; namely,
the ethical-religious. Their material, and thereby also their scientific, cul-
ture was borrowed from the earlier developed and powerful neighbouring
peoples and may therefore be passed over here. Nevertheless the Israelitic
conception of nature as preserved in the Old Testament, has, owing to reli-
gious causes, right up to our own day had a deeply significant influence. The
part played by the six days' creation as a co-determining factor even in purely
scientific explanations of the world is too well known to need close exami-
nation. Likewise the ordinances of the Mosaic law regarding clean and
unclean animals have had their great importance for the conceptions of
nature held by the Christian peoples, while even the well-known problem
of the ruminant hare is still today a subject of lively discussion in cer-
tain circles. And undoubtedly, even in the far distant future, the religious-
CLASSICAL ANTIQUITY, MIDDLE AGES 7
dogmatic currents of thought which have always had, and indeed always
will have, a powerful influence on the development of human culture will
receive guidance from this quarter.
Hindu and Chinese science
The civilized peoples of eastern Asia, the Hindus and Chinese, have like-
wise contributed very little of importance to the development of the science
of biology. Hindu science, indeed, especially in the sphere of mathematics,
reached a high standard, and the tendency to employ figures even in the
other branches of learning which this people cultivated is unmistakable.
Thus a Hindu work on medicine states that the human body has seven skins,
300 bones, 107 joints, 900 tendons, 700 blood-vessels, and 500 nerves. But
they had very primitive ideas as to the functions of these organs, and
similarly the various fluids and kinds of air which provide for the body's
renewal are of interest to them more from the numerical than from the func-
tional point of view. Chinese culture, again, has essentially occupied itself
with ethical and social problems. Chinese medicine has on the whole ad-
vanced little beyond that of primitive peoples, although certain isolated
instances of progress achieved — for example, smallpox inoculation — might
perhaps be traced back to the experiences of this people. Even pure zoology
has on the whole made no advance; as early as about a thousand years before
Christ mention is made of an imperial zoological garden, but the thorough
study of the causal connexion in living nature did not come within the
sphere of Chinese interest.
CHAPTER II
THE EARLIEST GREEK NATURAL PHILOSOPHY
The Greeks: creators of natural science
IF THE Babylonians and Egyptians thus succeeded in collecting quite a con-
siderable mass of individual facts of science, it was nevertheless left to the
Greek nation to deduce from these facts a consistently realized conception
of nature — not free from mystical and magical influences, it is true, but still
striving more and more after a natural explanation of the laws of existence.
There has been much speculation as to why it should be amongst just this
people, who were not only few in number, but were also politically divided,
that such a splendid development of human thought should have taken
place. The deepest cause is surely to be sought in the much discussed, yet
fundamentally so inexplicable national character, in the spiritual and cul-
tural disposition of the people. It may, at any rate, be worth while briefly
considering its manifestations in the social sphere, in order to gain some
idea of the external conditions of development under which free thought
was here able to expand.
The people of Greece, as is well known, never achieved political unity;
it remained divided into a number of small communities independent of
one another, consisting usually of a city with its surrounding country dis-
trict. Trade and shipping rather than agriculture were the people's main
source of income. Over-population gave rise to splendid colonizing activity
along the coasts of the Mediterranean; the colonies, which from the very
beginning were made independent of the mother city, adopted the latter's
institutions. A strong national feeling prevailed everywhere and was main-
tained by law and custom. Outside the boundaries of his own town the Greek
was a foreigner without rights, without the possibility of acquiring civic
privileges elsewhere, and with no prospect of winning the consolations of
religion. Religion was in fact as localized as the communities themselves;
every town had its own gods, which could be worshipped only by its citi-
zens and within its boundaries. Such a local form of religion was naturally
primitive and remained so even at the time when Greek culture was at its
zenith. It was just on account of this lack of a more highly developed re-
ligion, however, that free thought was able to develop as it did. Here there
was no priesthood, as there was in Babylon, Egypt, and India, to reserve to
itself alone the right to the higher learning and to ensure that its results
8
CLASSICAL ANTIQUITY, MIDDLE AGES 9
did not conflict with the ancient religious usages. The religious persecutions
that occasionally took place in Greece against thinkers, as, for instance,
against Socrates, were rather the work of the mob than of the defenders of
religion and were therefore of a purely incidental and transitory nature. On
the other hand, we find that many of Greece's oldest philosophers were
priests or at any rate the sons of priests. And just as religion in ancient
Greece was primitive, so also were the moral ideas: provided the citizen
obeyed the ancient laws of the State, he need not worry much about what
further duties were owed to his nearest and to himself. Thought was thus
at liberty to turn to external nature and devote itself to speculations on
how things arose and why the world and the living creatures in it were
formed just as they were. The oldest Greek thinkers were therefore natural
philosophers, while it was not till later that the ethical problems — which,
for instance, among the thinkers of the Jewish people, the prophets, had
from the very beginning dominated the soul — through Socrates found a
place in Greek thought and finally, in late classical times, entirely sup-
planted the interest in nature and its phenomena.
These, mankind's earliest natural philosophers, went about their work
under conditions which in most respects were utterly primitive. The general
education amongst their neighbours was extremely limited and far from wide-
spread — in fact, throughout the whole of the classical period of Greek cul-
ture it was confined to a very few. The public instruction provided by the State
for the benefit of its citizens was of the simplest kind; in Athens in the time
of Pericles, when the greatest philosophers and poets of Greece were as-
sembled there, the citizens had to learn in the State schools only the simplest
rudiments of reading, writing, and arithmetic, besides music and gymnastics,
which were necessary for military service, while it is said that at the same
period in the more conservative country of Sparta the majority of the peo-
ple were illiterate. Anything that the private individual wished to study
beyond that, he had to find out for himself as best he could. Nor were
there in ancient times any private professional teachers. If a person of studi-
ous mind happened to belong to a family connected with the priesthood,
its traditional learning was naturally at his disposal as a foundation; for
the rest he had to rely upon whatever knowledge he could acquire in his
own city from foreign travellers and such of his countrymen as had travelled
abroad, unless he himself was rich enough to travel and visit learned men
in their own homes. Fortunately hospitality in Greece in ancient times knew
no bounds; in actual fact it took the place of learned schools and universities
and even of books and writings. For if the knowledge of writing was rare,
this was to a great extent due to the difficulty of obtaining writing-materials.
The Egyptians had discovered a cheap material in their papyrus, the Chaldees
another in their clay tablets, but the ancient Greeks had nothing but metal
lO THE HISTORY OF BIOLOGY
tablets and animal hides, both of which were expensive to get and incon-
venient to preserve. The learned therefore had to express their opinions in
short and weighty compositions, preferably in the form of verse, so that they
could be easily learnt by heart. Thus learning became the asset of a privi-
leged few; they had to be wealthy in order to be able to undertake the
journeys that were essential for the acquiring of knowledge, and of high
standing in order to be able, both at home and abroad, to gain access to
the masters who were primed with the wisdom of the period. But in point
of fact the scholars of those days were highly respected: the various states
summoned them to be lawgivers and rulers, paid the expenses of their
costly journeys, and gave them financial assistance when they ruined them-
selves over their research work. On the other hand, they were often per-
secuted by hostile political factions and were sometimes condemned to end
their days in exile.
The earliest scientists of Greece: the Ionian philosophers
The earliest of these Greek natural philosophers, the so-called Ionic philos-
ophers, all lived in, or at any rate originated from, the colonies which
the Ionic tribes of Greece founded on the coast of Asia Minor. Through
trading with the Orient these cities rapidly grew wealthy, and through
contact with the more highly cultivated peoples of the East there arose a
keen desire for knowledge, and means for satisfying it were obtainable.
Chaldean and Egyptian travellers were able to tell of the great learning of
their priests and physicians; journeys to the East gave the ambitious lonians
opportunities of acquiring at least something of that secret knowledge.
And on this foundation they themselves built further. — Nature presented a
great number of phenomena in constant variation, and herein it was proved
that certain phenomena always stood in a certain regular relation to one
another. A common primary cause of the variations of existence had to be dis-
covered — a common element out of which everything originated. What was
this primary element and how have things originated from it? These two
questions occupied the minds of the Ionian thinkers. Nature, the Greek
(pixTLs, became the one great problem, and these ancient philosophers who
studied the problem of nature were therefore called physicists, a name which
later on was reserved for those who carried out research in a limited sphere
of natural science. The investigations carried out by these ancient physicists,
however, led them just as often into the realms of metaphysics, and it is
just that lack of insight into the insuperable bounds of natural science that
gave to their speculations that vague and fantastic character which is so
conspicuous in them.
As one of the earliest of the natural philosophers in Greece is mentioned
Thales of Miletus. Even in ancient times very little was known of his life
and activities The very epoch in which he and his immediate successors
CLASSICAL ANTIQUITY, MIDDLE AGES II
lived has been so variously stated that the dates differ by centuries. It is
most generally assumed, hov^ever, that he lived between 650 and 580 b.c.
Historians agree in declaring that he .left no writings; perhaps he was not
even able to write. He was probably of Phoenician origin — some assert that
he immigrated from Phoenicia. At any rate it is clear that he had educated
himself by travelling and studying in the East. He was very rich and of high
standing and collected around him a number of disciples. Of his philosophy
it is mentioned that he regarded water as the cause of all things. The earth
floated like a disk on a vast sea which surrounded it on all sides. The details
of his philosophy are unknown, but the assumption mentioned above is to
a certain extent reminiscent of the story of the creation in Genesis, with
its definite assertion of "waters which were under the firmament" and
"waters which were above the firmament." That we are here dealing with
a theory of oriental origin seems beyond all doubt. That Thales was the
pioneer of the Greek natural philosophy is undeniable; he is unanimously
acclaimed as such by the thinkers of antiquity. The very name of philosopher
and philosophy probably originates from him. Once asked whether he was
a wise man (cto^os in Greek), he modestly replied that he could not call
himself one; he was merely a lover of wisdom (0tX6o-o<^os) .
A younger fellow-countryman of Thales, and in all probability a disciple
of his, was Anaximander, who lived approximately between the years 611
and 546 B.C. Concerning his life and personality about as little is known
as of that of Thales. On the other hand, it is known that he described
the results of his scientific researches in a poem On Nature (jrepi (i>v(Ttb:s),
which is quoted by several later philosophers. Even Aristotle declares that
he had read it, but it seems to have been lost as early as the later classical
period; the people of antiquity had not such great respect for "classical"
authors as we have in our time. Through quotations and references in the
writings of later authors, however, it is possible to form some idea of this,
the first work on natural science ever written. Just as for Thales, the most
important question for him is: What is the material cause of the universe?
As mentioned above, Thales held that water was the causal principle;
Anaximander conceives it to be "apeiron" (aireipov), from which he
supposed the things on earth to develop themselves and into which he
supposed them to return. What he actually meant by this "apeiron" it is
difficult to say, but the word probably means " the quality-less, the indeter-
minate." Out of this primordial cause have arisen heat and cold, from these
water, and from that again earth, air, and fire, which last surrounds the
atmosphere and is radiated through the stars. The earth came into being
through a kind of condensation of water; it was originally composed of
pristine mud and then became solid and floats as such on the water, in form
like a spherical segment — that is, very much like a loaf. He is said to have
II THE HISTORY OF BIOLOGY
designed a map of the world and even to have made a celestial globe of spheri-
cal form, with the earth suspended in the centre of the circular vault. Living
beings he conceives as having evolv.ed through a kind of primordial pro-
creation in the mud which formerly covered the earth. Thus, first there
arose animals and plants, and then human beings, who, originally formed
like fishes, lived in the water, but afterwards cast off their fish-skin, went
up on dry land and thenceforth lived there. We see, then, that Anaximander
produced a complete theory of evolution, childishly clumsy, it is true, but
interesting for the audacity with which he deduced his conclusions from
his premisses. Nor is he afraid of letting the world-process continue into in-
finity; the present universe has been preceded by others, which were evolved
out of the primordial element and returned once more to it, and so it will
always continue. We should not go too far, however, in making a comparison
between the ancient Ionian's theory of creation and the evolution theory of
our own day. An attempt has been made to prove him a predecessor to Dar-
win on the ground of his above-mentioned conceptions of the origin of
man. This is an entirely unhistorical view of the matter, although an easily
accountable one; highly debatable theories have always sought for direct
predecessors as far back in time as possible. Anaximander's theory of the
origin of man is in reality most reminiscent of his fellow-countrymen's
legends of autochthonism — ■ stories of how men were born of the earth they
lived on, which was one of the very popular myths of these periods of migra-
tion of peoples, whereby they sought to justify their title to the country
they possessed on, as it were, semi-natural, semi-divine grounds. But if we
must exercise caution in gauging the speculations of Anaximander by modern
standards, they at any rate are worthy of our high admiration. Natural re-
search in our own day endeavours to discover an explanation, based on a nat-
ural connexion of causes, of the origin of things and of the variations they
present, and the philosopher who was the first to realize the necessity of
such a natural explanation and who worked it out, although incompletely,
must be regarded in all ages as one of the pioneers of human thought. The
religious-poetical myths of the creation, which up to that time had had to
serve amongst the Greeks, as in the East, for an explanation of the cosmos,
became from this time part of the sphere of poetry and the life of religious
faith, while scientific research went on building upon the foundations laid
down by Anaximander.
Among his contemporaries Anaximander enjoyed a high reputation, and
several of his disciples are mentioned in history. In his native city his work
was carried on by Anaximenes, who chose air for his principle and considered
that this not only enveloped the world, but also penetrated living beings
and represented their life-principle. Shortly after his death the city of Miletus
was ravaged by the Persians and razed to the ground (494 b.c), and with
CLASSICAL ANTIQUITY, MIDDLE AGES 13
that the city of Thales and Anaximander disappears for ever from cultural
history. Their theories, however, had been widely dispersed, and when the
Asiatic Greeks lost their cultural supremacy, philosophers and philosophic
schools were already to be found scattered throughout the world of Greek
culture.
Thus Diogenes of Apollonia, in Crete, is regarded as one of the Ionian
school of philosophy. He lived in the first half of the fifth century and is not
to be confused with the more famous Cynic Diogenes, who lived in the time
of Alexander the Great. His explanation of the universe is based on An-
aximenes' theory of air as the primary matter. Out of the air are formed all
other elements in the world through a process of condensation. He conceives
life to consist of warm air moving like currents through the veins and thus
preserving the strength in the body. Diogenes has described the ramifications
of the venous system in man, or rather in mammals, and this description is
still partially extant — the earliest anatomical work known. For the rest,
we know of Diogenes that he conceived living beings to have been produced
out of the earth through the influence of solar heat — that is to say, a fur-
ther development of Anaximander's theory. Also, he believed that the
embryo in the uterus was developed by the warmth of the mother out of
the semen of the father. His embryological statements must therefore, like
his anatomical ideas, have been based on dissection. A contemporary of
his was Hippo, who is also said to have engaged in embryological research.
Unfortunately we know very little about him; we are not even certain of his
birthplace, which some say was the Isle of Samos and others Rhegium, in
the south of Italy. His reputation as a philosopher seems to have been in-
ferior to his fame as a naturalist, which is largely the reason for his having
been almost forgotten. He is said to have maintained Thales' theory of
water as the origin of matter.
This survey of the old Ionic natural philosophy shows that there was a
serious attempt made to discover a natural connexion in the events of earth,
in the existence, origin, and decay of matter. However, partly through the
adoption and development of its ideas and partly through fresh influences
from the East, there grew up side by side with it other lines of thought, hav-
ing in some ways a deeper vision of the phenomena of life, but at the same
time also a tendency to mysticism and fanciful ideas which had been foreign
to the Ionian philosophers. Although it was not particularly interested in
biological research, it is necessary here to mention the Pythagorean philoso-
phy, owing to the important part it plays in the history of culture in general.
Its founder, Pythagoras, is one of the most extraordinary figures in cultural
history. Scientist, religious prophet, and statesman, mathematician, and mys-
tic all in one, he has become in the tradition of posterity a purely legendary
figure. Born in Samos off the coast of Asia Minor, he travelled widely in the
14 THE HISTORY OF BIOLOGY
East — how far is not known with any certainty — and afterwards taught
in his native island, but on account of political disturbances he was forced
to migrate to the Greek colony of Croton in the south of Italy. There he
carried on as a research-worker and religious and social reformer until his
death, probably about the year 500 b.c. His dates are in any case highly
uncertain and much disputed. There is no doubt that much of the wisdom
he taught emanated from the East; his famous theory of the wandering of
the soul, for instance, already existed in India long before his time, and the
geometrical theorem named after him had already been proved by Indian
mathematicians long before he lived. His cosmological theories, however,
are extremely interesting. He conceived fire to be the origin of matter. It
should be borne in mind that to the people of antiquity and to many suc-
ceeding generations fire was not a chemical process, but an element, like air,
water, and earth. Pythagoras believed that all things originated in a primor-
dial fire forming the centre of the cosmos. Around this primordial fire re-
volve all the celestial bodies, the earth, the planets, and the sun. The shape
of the celestial bodies is spherical and their orbits circular. This cosmology,
as compared with that of the lonians, represented an immense advance.
Through him the fact of the globular shape of the celestial bodies was in-
troduced into science, although it took a thousand years to penetrate the
consciousness of the world in general. Still more remarkable was his theory
of the earth as a moving, revolving body. This theory the people of antiquity
found it impossible to accept; ^ Copernicus was the first to take up the theory
anew and was actually accused by his opponents of Pythagoreanism. In
the sphere of mathematics Pythagoras was also a pioneer; he discovered the
regularity of number-series and was led by his speculations in mathematics
to propound the fanciful and mystical theory that numbers govern matter,
and even that numbers are the principle of matter. Further, he even included
tonal harmony in music in this mystical system of thought; his theories on
the "harmony of the spheres" are as well known by name as they are diffi-
cult to understand in substance.
Pythagoras' influence on scientific development was very great and was
also considerable in the political life of his day. His disciples founded a
strict order, or kind of sect, which worshipped in Pythagoras a divinely
inspired prophet. Persecuted during the most brilliant period of Greek
democracy for their pronounced aristocratism, they were finally dispersed,
but their teachings experienced a revival in late classical times.
"The Greeks of the West
With Pythagoras the nationality of western Greece assumes a place in
scientific history. Southern Italy as far as Naples, as well as Sicily, had been
* Aristarchus of Samos taught, it is true, at a far later period that the sun is the centre
around which the earth revolves, but this theory soon fell into oblivion.
CLASSICAL ANTIQUITY, MIDDLE AGES 15
colonized by the Greeks at an early period, and by the extermination or the
assimilation of the aboriginals a homogeneous Greek nationality had grown
up here, split up into small states, as in the mother country, and living, if
possible, under still worse conditions of political unrest. But their intellec-
tual culture rivalled that of the Asiatic cities. A peculiarity of the western
Greek philosophers was that a number of them were, like Pythagoras,
imaginative prophets and imperious statesmen, and at the same time keen
research-workers, while they founded schools of a far stricter order than
their Ionian predecessors. One philosophic school of this kind was the
Eleatic school, called after the city of Elea, in southern Italy. There came to
this city at the end of the sixth century b.c. a man whose name was Xenoph-
ANEs, born at Colophon in Asia Minor and a disciple of Anaximander.
Disturbances in his native city had driven him into exile and he had wandered
far and wide in the greatest poverty, supporting himself by reciting his own
poems in the towns he visited. Finally in Elea he found a place of refuge
and died there about 490 at a very advanced age. The results of his scientific
researches he has described in a poem, similar to his master Anaximander's
treatise On Nature. Some fragments of this poem are still extant. In spite of
the extraordinary audacity of the ideas which it contained, it won its author
a great reputation and a large number of disciples. He based his ideas on
Anaximander's theory of the origin of the world through the condensation of
water and primordial mud, and he developed it still further. Of interest in
this connexion is his pointing out fossilized marine animals high up in the
mountains, which he declared to be a proof that the mountains were at one
time under water. These ideas were neglected, mainly owing to the fact that
Aristotle and his disciples regarded fossilization as one of the "lusus natura."
It was not until the Renaissance that Xenophanes' more correct views once
more came into their own. But speculations as to the origin of the world
drove Xenophanes further and further over to purely theological problems.
He became a keen and eloquent opponent of his fellow-countrymen's belief
in a plurality of gods, which he despised on account of their purely human
limitations; horses and oxen, he declared, would, if they thought as men,
imagine gods in the form of horses and oxen. On the other hand, he for his
part maintained the eternity and unfathomableness of divinity, and con-
sistently therewith the eternity, unity, and immutability of the world in
which we live. This did not prevent him, however, from embracing Anaxi-
mander's theory of alternating evolution and annihilation of the earth and
all that lives on it. But it was the theory of immutability that his disciples
further developed. The most famous of these, Parmenides of Elea, vigorously
maintains the unity of being. The world is conceivable, he declared, only if
we disregard the variations and changes and seek the immutable. In this
connexion he warns us against relying upon the senses, whose judgment is
l6 THE HISTORY OF BIOLOGY
misleading and from which he appeals to reason. His abstract conception is
also seen in the antitheses by which he expresses existence: hot as opposed to
cold, light as opposed to darkness, the "ent" (being) as opposed to the "«<?«-
ent" (not-being). In regard to man, he conceived the soul, which to him was
the same thing as life, as hot and the body as cold. For the rest, he accepted
Anaximander's theory of the origin of living creatures and the doctrine of
the Pythagoreans as to the globular form of the celestial bodies. His great-
est service to mankind lies in his insistence upon logical consistency in
thinking; in this he far outstripped the Ionian philosophers and strongly
influenced the thinkers of the ages that followed. The later Eleatics finally
pursued the theory of immutability to sheer absurdity and thereby rendered it
untenable. The Eleatic Zeno, for instance, denied all change and even motion.
Of far greater importance for the development of biology than the
Eleatics, however, was another western Greek philosopher, Empedocles,
of Acragas, in Sicily. His period of activity is generally placed in the middle
of the fifth century. Around his personality and way of life there has grown
up, as around Pythagoras, a number of legendary tales, which prove, if
nothing else, that he must have very greatly impressed both his contempo-
raries and posterity. And this seems to have been very much his own inten-
tion. He boasts about himself in the writings, fragments of which have been
preserved to us, concerning his own supernatural gifts; he claims that he has
power to heal the sick and cure the infirmity of old age, raise the dead,
change the direction of the wind, and bring rain and sunshine upon the earth.
And he delights in being acclaimed; adorned with chaplets and flowers, he
goes in procession into the city that besought his help and is hailed by the
inhabitants almost with the reverence due to a divinity. In our days this
would, of course, be characterized as shameful humbug, but in early times
it was apparently not so. Undoubtedly Empedocles himself believed in his
miraculous powers, and the taste for pageantry he shared with his own
countrymen. History also relates a number of serviceable acts he performed;
for instance, he improved the hot and unhealthy climate of his native city by
making a breach in the mountain wall which shut out the cool north wind;
he rid a neighbouring town of malaria by arranging for the draining of the
district. He was, besides, a leading politician; although descended from a
distinguished family, he was a keen democrat; he overthrew the oligarchies
in his native city and set up a popular government; the honour of kingship,
which was offered to him, he declined. However, his enemies prevailed over
him and he had to flee in exile to Greece, where he died. Shortly after his
death Acragas fell into the hands of the Carthaginians and was razed to the
ground. True, the city again flourished in the time of the Romans under the
name of Agrigentum, now Girgenti, but the part it played in the history of
culture was at an end.
CLASSICAL ANTIQUITY, MIDDLE AGES 17
The four elements
As a philosopher Empedocles bases his theories on Parmenides. The world
is uniform, immutable. Nevertheless changes do take place, but these must
be explained by movements in existing matter and by the alternating com-
mixture and dissolution of its component parts. As the fundamental ma-
terial causes Empedocles postulates fire, air, water, and earth — in other
words, the four elements or roots. The theory of these elements, which has
been maintained, one is almost tempted to say, up to the present day, origi-
nates, as was universally acknowledged by antiquity, from Empedocles. And
if he asked what it was that produced the motions in the four elements which
caused the changes in their constitution, he would answer: love and hate;
love acts as attraction, hate as repulsion. Through their alternating predomi-
nance are produced all the changes in nature. Love, when it ousts hate, gives
rise to new worlds; when hate predominates, they are again dissolved. The
world was formed by parts of the four elements becoming united in a work;
through the same kind of motion the water elements were flung out of the
originally humid earth-mass, so that the latter became dry and habitable.
If fresh beings arise or such beings as have existed perish, it is not to be con-
cluded from this that something can arise out of nothing or become nothing;
for whence could anything new come to that aggregate of reality that exists,
or whither could anything already existing disappear? No, the cosmic matter
was and remains the same, only its component parts are mixed in different
ways through love and hate alternately predominating. Living creatures he
conceives as having arisen out of the earth; first, plants, whose life he com-
pares in detail with that of the animals; their nourishment is procured
through pores in the stem and leaves, and their germination is comparable
to the reproduction of animals. Animate organisms have likewise originally
sprung from the earth; first arose individual limbs and later, through the
powers of attraction of love, these were conjoined; from them then arose
animals themselves. But this development did not proceed undisturbed, for as
the conjoining of the separate parts took place by accident, it was entirely a
matter of chance whether beings capable of life or malformed monsters arose.
Mankind also originated in a similar way; through the co-operation of the
subterranean fire there were cast up out of the interior of the earth shapeless
lumps which formed themselves into limbs, from the union of which man
developed. Men, who are of a warmer temperament, came into being in a
southern climate, while women, the more cold-blooded sex, were created in
a more northerly climate. In reproduction the embryo received some parts
of the body from the father's and the others from the mother's seed. Growth
in childhood is due to increase of warmth in the body, and the infirmity of
age to its diminution. Respiration he believed to be effected not only through
the windpipe, but also through the pores of the skin; as the blood is conveyed
l8 THE HISTORY OF BIOLOGY
alternately to and from the skin, the air is inhaled in conjunction with it.
The perceptions of the senses are due to the objects that are perceived or
sensed giving off fine particles, which unite themselves to the corresponding
components in the organs of sense. Thus the various elements are apprehended
by corresponding elements in the organs of sense: water by water, air by air,
etc. Tones are created by the air brought into movement forcing its way into
the auditory duct of the ear as into a trumpet. Empedocles is said to have
been the first to describe the labyrinth of the ear. The eye he compares to a
lamp; light is distinguished by the eye's fire-components, darkness by its
water-particles. Even in the operation of thinking he sees a purely corporeal
function; the blood, in which all the elements are most minutely commingled,
is the seat of intelligence, but in this other parts of the body can also co-
operate; elsewhere in the operation of thinking the four elements in the
thinker and the thought seek one another. The more finely and evenly the
elements are mixed in a man, the better does he think; if the correct mixture
is confined to particular parts of the body, then these parts are more highly
developed than the others. — In curious contrast to this materialistic theory
of sensation are several utterances of Empedocles in which, like Parmenides,
he warns us that the evidence of the senses cannot be implicitly trusted; they
can deceive, while the reasoned thought is infallible.
We cannot here enter into a discussion of Empedocles* religious theories.
Like Pythagoras, he believed in the migration of the soul and likewise for-
bade his disciples to kill and eat animals. Even certain plants were in this
respect sacred. For the rest, it is difficult to reconcile his mystical religious
pronouncements and his miracle-working activities with his natural philoso-
phy, which so strictly emphasized the doctrine of causality. From a closer
knowledge of the conditions under which he lived we might well have
been able to explain the riddle; now he stands as a unique, strange phenome-
non in the history of biological research.
Empedocles' scientific influence
In this sphere he undoubtedly deserves a high place. In particular his specu-
lations on the constitution of matter and its changes are worthy of atten-
tion. While his predecessors and even his successors for long ages afterwards
had no other natural grounds of explanation to offer in regard to these
phenomena than motion in space, Empedocles comes forward with a kind
of doctrine of affinity, crude and clumsy, it is true, but nevertheless con-
taining within it the germ of a number of ideas which it was only possible
for far later ages to think out. And the fact that these ideas were not adopted
by his successors, that antiquity failed to produce any science of chemistry,
does not detract from their interest. His physiological speculations, naive
though they are, also give evidence of his keen powers of observation and
combination. In his curious theory of the creation of living organisms from
CLASSICAL ANTIQUITY, MIDDLE AGES 19
scattered parts brought together by chance an attempt has been made to see
a kind of primitive theory of selection. That is, of course, an exaggeration;
his anthropogeny gives even clearer evidence than Anaximander's of being
derived from the old legends of autochthones.
With Empedocles the western Greek school made their final and most
important contribution towards the history of biology. A couple of hundred
years later this people saw the birth of their greatest philosophical genius,
the physicist Archimedes, who lived to see the destruction of the liberty of
western Greece. He was indeed a specialist in his own sphere, which has no
place in this work. We return therefore to Asia Minor, whence sprang the
whole of Greek natural philosophy and where the Ionian succession was still
being carried on with achievements of great importance for its further
development.
From its point of departure in the Ionian philosophy Heracleitus of
Ephesus (about 510-450) developed an entirely new idea of nature. He was
one of the greatest philosophers of antiquity, called "the dark," on account
of both his obscure style of writing and his gloomy view of life. He came from
a distinguished priestly family, but resigned an eminent government post to
devote himself entirely to philosophy. His life, too, however, was disturbed
by political revolutions. As a thinker he seems to have been essentially auto-
didactic, and though his system is to a certain extent based on the Ionian
philosophy, mainly on that of Anaximander, it nevertheless bears an en-
tirely original stamp. In contrast to the Eleatics' assertion of the immuta-
bility of the universe, Heracleitus sums up his teaching in the proposition
that everything is mutable, that mutability is the essence of existence.
"Struggle is life" is a saying that comes from him; "All is flux" is another.
He regards fire as the causal principle of the cosmos. Everything has arisen
from a primordial fire, to which everything returns, for worlds arise and
perish alternately. Heracleitus saw divinity in the primordial fire. Fire is also
the soul of man; fire is inhaled in breathing, and its cessation is therefore
identical with death. Disease arises mostly through water, the enemy of fire,
predominating in the body; intemperance in drinking clouds the soul, since
the wine makes the soul humid; "The driest soul is the wisest," he expressly
states in his writings. His special biological investigations have not been
preserved. He is said to have dissected animals, but it is not known whether
he drew any fresh conclusions therefrom. The service he rendered to science
lies in his general view of existence; in his constant insistence on the modifi-
cation of the principle, and at the same time on incontrovertible natural law
governing the universe — these two factors being the essence of existence.
In this he exerted great influence upon the natural philosophy of succeeding
ages, particularly upon Plato and, through him, upon Aristotle.
Contemporaneous with Heracleitus, however, there appears another
lO THE HISTORY OF BIOLOGY
line of thought which is far more concerned with a close study of nature and
makes it the basis of the entire cosmic system — the atomic theory. The
founder of this philosophy is said to have been Leucippus, a thinker of whom
we know nothing except that he was the teacher of Democritus, one of the
foremost natural research-workers and natural philosophers of all time.
Democritus was born at Abdera, a Greek colony on the Thracian coast. In
his time Abdera was a rich and powerful city, but in the course of subsequent
centuries it declined and its citizens became notorious for their stupidity,
which gave rise to the epithet "Abderitic" as a universal expression for
extreme foolishness. His period of activity is, like that of most of the early
Greek philosophers, not known for certain, but it is generally assumed that
he was born between the years 470 and 460 b.c. and died at a very advanced
age — in fact, not far short of a hundred years. From his father, one of the
richest and most eminent citizens of his native town, he received a large
inheritance, which he is said to have spent entirely on long journeys under-
taken for the purpose of acquiring knowledge from various countries and
peoples. On his return home he was supported by a brother until his fellow-
countrymen, proud of his scientific fame, granted him a pension sufficient
for all his needs. Known for his mild and friendly disposition and surrounded
by admiring disciples, he grew old in peace and died without suffering from
ill health. He was known as one of the most productive scientific authors of
antiquity, and his writings seem to have embraced many and varied subjects.
Except for a few fragments, however, they are entirely lost and indeed seem
to have already been so as early as the late classical period. Through other
ancient authors, however, particularly Aristotle, who always mentions him
with respect and seems to a large extent to have made use of his learning,
we are able to form a fairly good idea of his scientific point of view.
Maferialisfic theory of the universe
His teacher, Leucippus, seems to have based philosophy on Parmenides'
theory of the immutability of matter, and to have come across this in the
paradoxical form that it assumed among the later Eleatics. In order to pre-
serve the elements of truth in this theory and at the same time to make pos-
sible the changes which are incontestably observable in the world, Leucippus
conceived the universe as composed of a quantity of particles moving in
empty space. Democritus adopted this theory and developed it further. He
thus became the founder of the atomic theory, one of the most fruitful ideas
in natural science. And on this atomic theory he, as no one else has done
either before or since, based the whole of his theory of existence, both
spiritual and material. No thinker of antiquity ever produced such a consist-
ent and materialistic theory of the nature of the cosmos; none has advanced
further in the endeavour to satisfy the demand — upon which Anaximan-
der had already insisted — for a natural explanation of the origin of matter.
CLASSICAL ANTIQUITY, MIDDLE AGES II
From the fragments of his writings a number of general principles have been
gathered which are characteristic of his point of view; some of them sound
surprisingly like modern natural science.
Out of nothing comes nothing; nothing which is can be reduced to
nothing. All change is merely an aggregation or separation of parts.
Nothing happens by chance or intention, everything through cause and
of necessity.
There is nothing but the atoms and space; all else is an impression of the
senses.
The atoms are infinite in number and shape. Their movement is eternal;
in endless space they impinge upon one another, thereby producing vortices
out of which worlds arise, only again to perish. All the qualities of matter
are due to the shape, size, number, and motion of the atoms of which it is
composed. The atoms possess no other qualities than those mentioned above.
The soul consists of the finest and most mobile atoms; they fill the body and
give it life; if they leave the body, death ensues. Fire likewise consists of
small mobile atoms. On the whole, Democritus seems to have shared Hera-
cleitus' idea of the mutual affinity of the soul and fire. The stars he considered
to be like the earth, but owing to their rapid motion they were fiery bodies.
He seems also to have observed the mountains in the moon.
Biological knowledge of Democritus
Democritus has achieved important work in the science of biology and he is
here, as in many other directions, the finest of Aristotle's predecessors; how
much of his learning his successor borrowed we do not know, but there are
grounds for supposing that it was far more than posterity has ever guessed.
Without doubt he performed dissections of both the higher and the lower
animals and was the first to differentiate between them according to the
quality of the blood. The distinction between sanguiferous animals (verte-
brates) and bloodless animals, which was the principle of classification
adopted by Aristotle, originates from Democritus. In contrast to Aristotle he
believed that even the minutest animals possess perfected organs, although,
owing to their transparency, they are invisible to the human eye. In the
embryonic development the external organs arise first and the internal after-
wards. Many of Democritus' ideas we know only through Aristotle's po-
lemics against them, and in this respect modern research not seldom proves
Democritus right. For instance, he considers that the spider's web is pro-
duced from inside its body while Aristotle maintained that it is cast-off skin.
The sterility of the mule he seeks to explain by assuming a contraction of its
uterus. The construction and functions of the human body, however, were
naturally the main object of his studies. He conceives man to be a world in
miniature, a microcosm in ^ which every kind of atom is represented. He re-
gards the brain as the organ of thought, the heart as that of courage, and
ZZ THE HISTORY OF BIOLOGY
the liver as that of sensuality. His estimation of the brain places Democritus
once more in front of Aristotle, who believed the brain to serve only the
purpose of cooling the blood. Democritus considered life and the soul to be
one and the same thing, and the latter he believed, as already mentioned, to
consist of fire atoms which, owing to their lightness and mobility, are con-
stantly being given off by the body. Through inhalation the body receives a
fresh supply of them; if respiration ceases, then life departs from the body.
Sleep and asphyxia he declared to be also due to a loss of soul-atoms, but on
a smaller scale. Hydrophobia in dogs and human beings he considered to be
caused by inflammation of the nerves. Epidemics he believed to be the result
of atoms falling upon the earth out of other celestial bodies. Sensation is due
to the movement of atoms which emanate from the objects perceived. In
connexion with Democritus' materialistic idea of the soul he believed in
spiritual beings and revelations, a belief which other thinkers who regarded
the soul as matter — ■ as, for instance, Swedenborg — have shared with him.
On the other hand, he denied the divine beings of popular belief, without,
however, like Xenophanes, substituting any unified and eternal divine power
in their stead. Necessity, which, according to his views, governed the uni-
verse, was purely impersonal.
Democritus on the whole represents the climax of the endeavour of
Greek philosophy to arrive at an explanation of existence based on a natural
connexion of causes, an endeavour which, with Anaximander as its instigator,
gave rise to a long series of heterogeneous explanations of the cosmos, of
which only the most important can be considered here. In several funda-
mental respects — for instance, in the strict theory of causation, the atomic
theory, and, in connexion therewith, the principle of motion, the emphasiz-
ing of the importance of the brain for the function of thinking, the insistence
upon the complicated organism of the lower animals — Democritus achieves
results similar to those that have been attained by natural research in our
own day, although his speculations were in detail often very primitive, even
when compared with the achievements of philosophers of later antiquity.
The promising idea was not pursued, however, by succeeding generations;
for certain reasons which will be more clearly explained when accounting
for Aristotle's theoretical views, shortly after the age of Democritus the
Greek natural philosophy started to work out a method of explaining the
cosmic process entirely different from that indicated by the atomic theory.
Democritus thus represents the close of the first and the purely natural
scientific period of Greek philosophy. The results achieved by research dur-
ing that period have been mentioned above; what it failed to achieve may
also be briefly described here. The most serious defect from which it suffered
was undoubtedly the lack of material for investigation, which rendered it
difficult to follow up the principle of causation which had already been
CLASSICAL ANTIQUITY, MIDDLE AGES i}
determined. It should be remembered that antiquity knew nothing whatever
of chemistry, so that the ideas of chemical association and affinity were en-
tirely lacking as a basis for the changes in nature, and for these ideas the
vague and dogmatic theory of vortical motion was of course a very poor
substitute. Generally speaking, there did not exist in antiquity the ideas of
force and energy in the modern sense, but philosophers had to be content
with motion as an explanation of all changes. And, finally, they drew no
definite line between objective facts and subjective opinions; they had no
idea whatever of hypothesis. The whole of Democritus' cosmology was
purely dogmatic and was condemned to give way to other similarly dog-
matic explanations of the universe, which, though more attractive to his
contemporaries, nevertheless proved fatal to the future development of
natural science.
Reaction against natural philosophy
The first sign of the coming reaction is given by the philosophy of Anaxag-
ORAs. This thinker was a contemporary of Democritus, probably born
somewhat earlier than the latter, at Clazomenas in Asia Minor. His activi-
ties, however, were pursued in Athens in the time of Pericles — thus in the
community which proved to be the first great power of Greek nationality.
This is the first time in Greek scientific history that the town is mentioned
which was afterwards to become for nearly a thousand years the centre of
thought of classical antiquity. Athens was a city with an extremely mixed
population and an equally diverse intellectual life: side by side with the most
daring novelties in the region of thought dwelt crass superstition and fanati-
cal intolerance — to which latter Anaxagoras fell a victim. Accused of
"godlessness," hewas first cast into prison and then had to flee from Athens.
Nevertheless, his philosophy, as compared with that of Democritus, must
be regarded as idealistic. He conceived that the driving force in the universe
is what he calls the cosmic reason or the cosmic soul; a kind of spiritual
power to which he ascribes unity, omnipotence, and omniscience. This same
power is part of all living creatures and in them represents life itself. Matter
he believed to consist of an endless number of primary elements — that is, in
the same sense as that in which the old lonians conceived this idea. He does
not appear to have gone in for biological research, nor do his natural-scientific
views in general show any advance over those of Democritus. But he was
not without influence upon the succeeding ages and is therefore worthy of
mention.
The Sophists
Of far greater influence on the general course of development, however,
was a school of philosophers which appeared contemporaneously with him
and which began to lead Greek thought along entirely different channels.
These were the famous Sophists, whose founder and leading personality was
Z4 THE HISTORY OF BIOLOGY
Protagoras, likewise a contemporary of Democritus. Protagoras introduced
scepticism and subjectivism into Greek thought: "Man is the measure of all
things" is one of his fundamental principles; "Contrary assertions are equally
true" is another. To such a thinker any consideration of nature was foreign;
what he desired to teach was an art of living, such as would free mankind
from the fetters which traditional ideas in the sphere of religion and morality
had imposed. Highly acclaimed by the younger generation, bitterly hated
by the representatives of the past, not least because, in contrast to the earlier
philosophers, he taught for payment, which was at that time regarded as
equivalent to usury, Protagoras starts an entirely new era in the Greek view
of mankind, in relation both to his fellow-creatures and to nature. But be-
fore going on to describe this new tendency we must make a brief survey
of Greek medical science, at that period a specialized science which had
achieved results of lasting value to general biological development.
CHAPTER III
THE EARLIER PHASE OF GREEK MEDICAL SCIENCE AND
ITS SIGNIFICANCE FOR THE DEVELOPMENT OF BIOLOGY
Magical beginnings of Greek medicine
THE EARLIEST BEGINNINGS of the scicncc of medicinc among the Greeks
were as always in primitive medical practice, based on magical
religion, ^sculapius, the god of the healing art, had a numerous
priesthood, in which secret knowledge of the forces of nature and their use
for the curing of disease was jealously guarded and handed down from gen-
eration to generation. Pilgrimages were made to the temples of ^sculapius
by crowds of sick persons ■ — • both those in real and those in imaginary ill
health — to be cured of their maladies; and the latter class, the hypochondri-
acs, were not slow to spread abroad and confirm stories of the most wonderful
miracles of healing performed there. Immense hospitals were erected in the
neighbourhood of these temples for the benefit of those who required lengthy
treatment, and the very necessity of having to watch over and follow the
course of these patients' diseases must naturally have created a large
supply of purely empirical observations of immense value to the sacer-
dotal miracle-workers, who were thus enabled to estimate the result of their
methods of treatment. In time there grew up, as a result of the widespread
adoption of these empirical observations and methods, a class of purely
secular'healers, no longer directly associated with the temples of i^sculapius,
who nevertheless, in order to denote the origin of their art and to take ad-
vantage of the confidence inspired by religious belief, called themselves
Asclepiads — that is, descendants of i^sculapius. As other professions did at
that time, they formed a private guild, the members of which taught their
art preferably only to sons and kinsmen. Outsiders too were able by paying
large sums to obtain an insight into the secrets of the profession, while the
children of the family of Asclepiads always had the right to receive free
instruction from any of their father's colleagues. The wording of the oath
which the young physician had to swear before being allowed to begin the
practice of his profession is still extant. By this oath he pledged himself to
help teachers and his professional colleagues, give free instruction to their
sons, share with them any fresh experiences and discoveries, to the best of
his ability heal the sick, and refrain from mixing poisons and producing
abortions.
2-5
z6 THEHISTORYOFBIOLOGY
Earliest medical xvritings
The oldest known medical writings date from a period when medical science
was still materially, if not also in its ideals, dependent on the temples of
y^sculapius. There were in particular three famous shrines of the god from
which profane medical science was thus derived, and these were situated on
the islands of Rhodes, Cnidus, and Cos, all near the coast of Asia Minor.
The situation of these places shows, as did also the first nurseries of philoso-
phy, that oriental influence had been at work, and this influence did in fact
prove of incalculable importance for Greek medical science; particularly in
Egypt the art of healing had from the very earliest times been highly de-
veloped, and in historical times, too, was highly reputed even amongst
foreign peoples. Of the Greek temple schools mentioned, that on the Isle of
Rhodes was the earliest, that at Cos the most famous. The Coan school is
principally indebted for its fame to the family of Asclepiads originating from
there, which gave the world one of the greatest pioneers of medical science
known to history — namely, Hippocrates.
History has preserved the memory of seven Greek physicians called
Hippocrates; the one here in question is generally spoken of as Hippocrates
the Second or the Great. He is believed to have lived between the years 460
and 377 B.C. — that is to say, at the time of Democritus. He was born in the
Isle of Cos of the family of the Asclepiads which for several generations had
been attached to the temple of ^^sculapius on the island. He received his
medical education from his father, Heracleides, who, however, evidently
died when his son was still a youth. The young Hippocrates then betook
himself to Athens, where he studied philosophy with the Sophist Gorgias of
Leontini, and afterwards made several journeys in the Balkan Peninsula and
Asia Minor, eventually settling down in Thessaly, where he established a
large practice and finally died in the city of Larissa. His sons and grandsons
were likewise doctors whose reputation was high, though not comparable
to his own.
Hippocrates' fame as a pioneer of medical science is based chiefly on his
medical authorship. Even in ancient times, however, there prevailed some
uncertainty as to whether the writings that go under his name are genuinely
his, and in our own time only a few treatises out of what is called the Hippo-
cratic Collection can be accepted as coming from his own hand; the rest are
supposed to be partly the work of his school, partly Asclepiad writings dat-
ing from before his time, partly, and m.ostly, essays by considerably younger
authors. Hippocrates' own main treatise, Airs, Waters, and Places, con-
tains extraordinarily brilliant observations on climatical and geophysical
conditions and their influence on mankind in sickness and health, in their
material and spiritual aspects. The treatises in the Hippocratic Collection
dealing with anatomy and physiology, which are thus of interest for the
CLASSICAL ANTIQUITY, MIDDLE AGES Xy
history of biological research, are all believed to be of later date than Hip-
pocrates and are at any rate influenced by his views. They give evidence of
a close study of anatomy and physiology with the help of dissections and
vivisections. From the enunciation of the anatomy of the human body, how-
ever, it is clear that it is not based on dissections of the dead body, but that
the results of experiments carried out on the dead bodies of animals have been
applied to the human body without further evidence. To dissect the human
body, to violate the dead bodies of human beings, was from the very earliest
times considered a dangerous procedure and on that account was forbidden;
the reason for this was undoubtedly the universal fear of ghosts, which has
already been touched upon. It is certain, at all events, that, except in a few
particularly unprejudiced cultural epochs, it has always been difficult, even
in more modern times, to procure dead bodies for purposes of scientific inves-
tigation and to obtain permission to utilize them. The first part of the human
body to be studied with any great care was the bone-construction, which
could be examined in the skeletons of long-decomposed and consequently
less dangerous individuals, while the ancient custom of preserving corpses
by the preservation of the skeleton — a practice known in most countries
from prehistoric times — must have given cause for observing and getting
to know the various bones of the body. The musculature also, at any rate
its external layers, which it was possible to study in living human beings in
wrestling and athletics, has been comparatively well understood, while the
internal organs — the digestive, respiratory, and circulatory systems — re-
mained longest shrouded in obscurity, as regards both their construction
and their functions. These facts have also had an inflluence on surgery; while
fracture's and sprains were carefully studied and cleverly treated even in Hip-
pocrates' time, the art of arresting hemorrhage was extremely primitive;
from antiquity to the Middle Ages there existed no better method than cau-
terizing with red-hot iron, and it was not until the sixteenth century that
people learnt how to apply ligatures when operating.
Hippocrafic physiology: the four temperaments
The Hippocratic treatises assume with Empedocles and his successors that
the human body is composed of the four elements: fire, air, water, and earth.
To these elements correspond four "juices" in the body: blood, phlegm,
yellow bile, and black bile. Of these the yellow bile is produced in the liver,
and the black bile in the spleen. Proof of the existence of these juices was
found in the condition of the blood on coagulation, when, as is known, its
component parts become separated; in the undermost, black part of the clot
was recognized the black bile, in its uppermost, red part the blood; the
yellow bile was seen in serum, and the phlegm in the fibrin.^ The condition
^ See Fahraeus: The Suspension Stability oj the Blood (Stockholm, 1911).
x8 THE HISTORY OF BIOLOGY
of the body was due to the existence and commingling of these four primary-
elements; if they existed in the proper proportions, health was the result;
if the harmony between them was disturbed, sickness followed.^ For the
rest, Hippocrates and his successors seem to have shared the conception
which originated in Heracleitus that the soul, the life -principle, consists
of fire or — in later writings — of substance akin to fire, called pneuma
(breath).
In regard to human anatomy, osteology was, as has been mentioned,
comparatively carefully studied. The skull in particular was radically in-
vestigated and a large number of the names of its bones and sutures are
derived from these works. The bones of the face were also minutely studied.
The knowledge of the backbone was more defective, while, on the other
hand, the bones of the extremities were well described. Of the muscles sev-
eral, particularly the muscles of the extremities, are correctly described,
though the actual muscle substance is not accurately distinguished from a
number of other internal organs. The separate parts of the digestive canal
are named, but their connexion and function were extremely vaguely known.
The various sections of the intestinal tube are given different names and char-
acteristics in the different writings on the subject. The liver is an organ
which particularly interested the people of antiquity, as also the spleen, but
there were very confused ideas of their respective functions. A good deal
was known about the glands, especially the lymphatic glands; the pancreas,
on the other hand, was unknown. The function of the glands was believed
to be to segregate water from the body. Of the respiratory organs the larynx
and the trachea in particular were completely described, while the lungs
were treated only summarily. Respiration is believed to serve the purpose
of cooling the heart. Again, in another treatise it is asserted that the inhaled
air spreads to the various parts of the body, to the brain, the body cavity,
and the arteries. The enunciation of the circulatory system is, as mentioned
above, vague. The different cavities of the heart are fairly accurately de-
scribed, but here exists already the delusion which it has since been so dif-
ficult to eradicate that the left cavity of the heart does not contain
blood, but some kind of airy substance, which is proved by reference to
slaughtered animals, in which the arterial blood is drained off by severing
the jugular veins! The apprehension of the venous system is far vaguer than
that of the heart, and, moreover, the description of the former in the differ-
ent Hippocratic treatises is highly contradictory. The right side of the heart
2 The doctrine of the four temperaments, which counts for something even in our own day,
is based originally on the theory of these four juices and their distribution in the body; in san-
guine people the predominating factor is the blood, in phlegmatic people the phlegm, in choleric
people the yellow and in melancholy people the black bile. Melancholy, owing to its origin in
the black bile, is called spleen or atrabiliousness.
CLASSICAL ANTIQUITY, MIDDLE AGES 19
is the Starting-point of the circulation of the blood; the blood flows to it
from the rest of the body, and there obtains warm temperature from the
left cavity — that is to say, the blood coming from the body is "cold."
The left heart-cavity gets warmth from the air through the pulmonary veins.
The arteries likewise contain this warm kind of air, called pneuma, which
maintains vital action, and they disperse it throughout the body. It is truly
remarkable that, in spite of observations made from innumerable dissections
and vivisections performed by various research-workers, this primitive
butcher's experience of the emptiness of the left heart-chamber and of the
arteries should have been maintained up to the final era of the science of
antiquity.^ The warmed-up blood is forced from the right cavity of the heart
out into the body. In contrast to these primitive conceptions, however, cer-
tain of the Hippocratics have had some idea, though a vague one, of the
movement of the blood as an actual circulation.
Of the nervous system the Hippocratic school had, if possible, a still
vaguer idea than of the circulation of the blood. The brain was believed to be
a gland which segregates water and possesses the function of cooling the
blood and collecting mucus out of the body. This mucus is then segregated,
together with the water, by a catarrhal affection through the nose. Certain
later Hippocratics, however, probably influenced by Democritus, have a more
accurate view of the functions of the brain, believing it to be the centre of
thought, feeling, and motion. The nerves are invariably confused with the
tendons, sometimes also with the veins, and for this reason all ideas of the
functions of the nervous system are already ruled out. Certain of the most
important cerebral nerves are, however, described and named. The construc-
tion of the eye was fairly thoroughly studied; its membranes and fluids, as
well as the pupil, were known, but the lens was unknown. Sight was pro-
duced as a reflection of the object seen on the pupil. Of the ear the bony laby-
rinth, the auditory canal, and the tympanum were known. The urogenital
apparatus is described in its main features; regarding fertilization there
existed then, as indeed throughout antiquity, extraordinarily fantastic ideas.
With zoology proper the Hippocratics naturally had little cause to con-
cern themselves. Nevertheless, there exists a treatise On Diet, in which there
are enumerated fifty-two different edible animals, arranged on a certain defi-
nite system; first quadrupeds, tame and wild, birds, fish of several kinds, in-
cluding coast-fish, mud-fish, river-fish, mussels, and crayfish. This so-called
Coan animal system has the advantage of differentiating between various cate-
gories of living creatures — a first primitive attempt at proper systematization.
3 This idea possessed, it is true, the sanctity of religion; in sacrificial animals the arteries
were naturally empty, and the interior parts of these animals were examined with a view to
basing on them prophecies of the future. To deny the results of such examinations would of
course have involved wounding time-sanctioned religious susceptibilities.
CHAPTER IV
THE END OF N A T U R A L - P H I LO SO P H I C A L SPECULATION. THE
PREDECESSORS OF ARISTOTLE
The results of natural philosophy
A N ENTIRE ERA in the history of biology closes with the philosophers
/ \ that have here just been characterized — Democritus, Hippocrates,
JL JL and his school — an era which may properly be called the era of
natural-philosophical speculation. The results achieved by their researches
cannot be regarded as anything but magnificent; for the first time in the his-
tory of humanity to have built up a real natural science is an achievement
worthy of the highest admiration, however modest the results may have been
in certain details. This research work had so far been directed by three illus-
trious representatives, Anaximander, Empedocles, and Democritus. All of
them sought an explanation of existence as a natural course of events; in
this direction Democritus proceeded as far as human thought has at any
period proved capable of going. Nevertheless, even in his own lifetime the
revolution was being prepared which was shortly to lead Greek thought in
entirely different directions. The first ideas on which this change was based
originated, as hinted above, from the school of the Sophists. The new prin-
ciple that they taught was subjectivity: "Man is the measure of all things. "
To this really true assertion the ancient "physicists" could make no objec-
tion; indeed, their cosmic explanations were as numerous as themselves,
and each one of them could only declare dogmatically that his own views
were the true ones and thereupon produce a number of more or less illogical
arguments in support of them. The claim of the Sophists as to man's being
the measure of all things thus for the time did away with all objective ex-
planations of natural phenomena, for what was the use of disputing about
matters which all viewed from different standpoints if all could be equally
right? Sophistry itself, however, when consistently applied, led to pure
nihilism, both intellectual, in that the object was, by means of ingenious
turns of phrase ("sophisms"), to prove absolutely anything, and moral,
in that all generally accepted sound traditions were held in contempt as
laying a restraint upon the individual's freedom of action. If all scientific
thought were not to be destroyed altogether, its preservation must be sought
by turning the whole trend of thinking into quite a different direction. And
this was found by maintaining that human thought, however much it may
3°
CLASSICAL ANTIQUITY, MIDDLE AGES 31
be the measure of all things, such as they appear here on earth, is nevertheless
itself subject to eternal laws, more infallible than those which the old phys-
icists saw in existence. The men who thus saved the Greek philosophy from
degenerating into empty rhetoric and worthless quibbles were two Atheni-
ans, Socrates and his disciple Plato. Socrates worked exclusively in the
ethical sphere; in this he sought for standards binding for all and emanating,
not from ancient tradition, but from the conscience of the private individual.
"Anyone can become virtuous if only he accepts a knowledge of virtue";
that is his principal doctrine, and this knowledge he for his own part derived
from a divine voice in himself, which he desired also to awaken in his fellow
human beings. Nature did not interest him in the least; the streets of Athens
were his haunt, he said, and neither trees nor stones had anything to teach
him.
Plato, a disciple of Socrates, generalized the doctrine of standards in
the ethical sphere which he learnt from his master, so that it was applied to
embrace the whole of the intellectual life of man. Born at Athens in 4x9 of
a distinguished family, he attached himself to Socrates at the age of about
twenty. Upon his master's death he left Athens and made extensive journeys,
afterwards returning to Athens, and there he established a school or college
known as the "Academy," which survived long after his death. He died
in 347. Like Pythagoras, he was a clever mathematician and, also like him,
combined ah inclination for the conclusive deductions of mathematics with
a strong attraction for the mystic. In the dialogue Thnaus, in which he pro-
pounds his theory of the origin of the universe, the functions of the human
body, and the relation of man to nature, he has evolved a history of creation,
poetically very fine, but at the same time purely mystical, testifying, it is
true, to his high ethical aims, but of no greater value as natural science than
any of the ancient popular cosmogonical myths. The world was created by
an eternal and perfect god, and therefore there can be no question of an endless
number of worlds, as Anaximander and Democritus made out, but only one,
and this single world must have received the most perfect of all shapes, the
sphere. He accepted Democritus' atomic theory to the extent that he believed
matter to be composed of particles, though these again are not of endless
variety, but are five in number, corresponding to the five regular polygons
of the geometry of space (Plato was one of the founders of this science and
one of the foremost geometricians of all time) — so that each element has
its atomic form: fire the pyramid, the earth the cube, the air the octahedron,
and water the icosahedron; the dodecahedron represents the heavens. Now,
this geometrical atomic form is in itself no more dogmatic than the entire
atomic theory of Democritus, but it forms the starting-point for a process
of natural speculation which is diametrically opposed to it. For while
Democritus takes as his starting-point the atoms and the matter formed by
31 THE HISTORY OF BIOLOGY
them as being the only really existing element, Plato finds true reality in
the world of abstract thought and maintains that what is perceptible by
the senses is an imperfect image of the eternal ideal, the divine intelli-
gence, conceivable only by abstract thinking. The nearer things are to
the divinity, the more perfect and animate do they become. Thus the
stars of heaven possess a higher animate life than man and are created in
greater likeness to the supreme intelligence. Of man the first thing to be
created was the head, which Plato, like Democritus, regarded as the organ
of the animate soul; the head has very nearly the spherical form corre-
sponding to the ideal, and trunk and limbs are created to save the head the
trouble of rolling along the ground. In the trunk dwells a lower, mortal
soul, whose best part, the heart, the organ of courage, is separated by the
diaphragm from the organ of animal desires, the digestive apparatus, the
most vital part of which, the liver, however, has the merit of producing
dreams during sleep, from which the future can be predicted. The plants on
our earth are created to provide man with food; animals, again, are sprung
from men whose souls have degenerated and have consequently been given
an inferior dwelling; first, women have come into being out of cowardly
men's souls, then birds and quadrupeds out of such human beings as have
neglected their intelligence, and, finally, the most worthless souls have been
placed in aquatic animals, "which may not even breathe pure air." This
theory of the migration of the soul is clearly reminiscent of that of Pythag-
oras; from him, too, no doubt originates Plato's mystical theory of number,
the details of which do not belong to our present subject.
Systematt%ation of thought
It may seem unnecessary to dwell so long on these fantasies. They are, how-
ever, well worth noting, for their originator has exercised a rare and radi-
cal influence on human culture in its entirety and even, as we shall find later
on, no small influence upon the development of biological science. His philo-
sophical speculations related above have in this respect had but little signif-
icance. Plato's greatest contribution has been made in the sphere of the
purely ideal intellectual life. The spirit of man is, he said, bound by laws far
more abiding than those that may be deduced from natural phenomena. To
these inflexible standards for intellectual activity he is led by speculations
in the ethical sphere, in which, following Socrates, he found a definite paral-
lel between human actions and their consequences. Having advanced thus
far, he devoted himself to developing this theory of the conditions, bound
by immutable law, of the world of ideas, in which, as mentioned above, he
saw the true existence, of which the phenomena visible in natural life are
mere images. Everything on this earth, then, has its eternal idea as its proto-
type; every individual horse is an imperfect image of the idea horse, which
is eternal and perfect. Through this reasoning Plato came to be the founder
CLASSICAL ANTIQUITY, MIDDLE AGES 33
of the system of ideas, which has played an important part in biology; out
of the idea horse in contradistinction to the individual has arisen the no-
tion of species, and gradually likewise all higher systematical categories. It
is easy to realize the immense influence this has had on biological research. A
mass of detailed ideas in systematization — for instance, the dichotomic classi-
fication tables of genera and species — come from the Platonic school. And,
mathematician as he was, Plato further endeavoured to make his inferences as
conclusive as possible; the listener and the reader should realize that the
result of the investigation must be so and could not be otherwise. Thus for the
first time in the history of science he not only made assertions, but also sub-
mitted proofs of them, based, it is true, upon abstract reasoning, as was the
entire world of thought which he built up, but at least formally convincing.
Theory of ideas
But if Plato, in laying the foundation of biological systematization, made
a powerful contribution to the progress of biology, nevertheless his activities
proved in other respects unfortunate for it. The enunciation of the world of
ideas as the true essence of being led to the underestimation of nature and of
the senses by which man observes nature. Plato realized the relativity and
limitation of observation through the senses,^ but not the arbitrariness to
which abstract thinking may lead if it is not controlled by observations. In
the Timaus he expressly states that no true knowledge is to be acquired
through the observations of the senses, but only a pleasure to the eye suitable
for a diversion — a statement which has been repeated after him by innumer-
able idealistic philosophers, both major and minor. Plato, the creator and
perhaps also, so far, the greatest upholder of idealistic philosophy, is like-
wise responsible for the contempt with which this trend of thought, which
otherwise deserves such high praise for contributing towards the develop-
ment of human thought, regarded natural philosophy. History indeed shows
that the more the idealistic philosophy governed man's desire for knowledge,
the greater became his indifference to the study of nature. In a far greater
degree than the traditional religions, idealistic philosophy has been the an-
tipodes of natural science.
On the whole, Plato's disciples followed in their master's footsteps.
Of the exact sciences mathematics interested them most and they worked at it
with great energy; otherwise, like many other schools of thought emanating
from the Socratic circle, they mostly occupied themselves with ethical ques-
tions. Even in Plato's lifetime, however, one of his disciples in the Academy
had begun to discard his ideas in a number of essentials and started to guide
philosophy in a new direction. This man was Aristotle, the greatest biologist
of antiquity and one of the most many-sided natural philosophers of all time.
^ As a matter of fact, Democritus had already realized this, but was unable to develop the
thought further.
CHAPTER V
ARISTOTLE
A RisTOTLE was bom in 384 b.c. at Stagira, a small Greek colony on the
/ \ Macedonian coast. His father, Nicomachus, belonged to an old fam-
^ 3L ily of the Asclepiads and was, like several of his ancestors, body-
physician to the Macedonian royal family. His predecessors among the Greek
philosophers had lived amongst the mobile and restless communities of the
city republics, while Aristotle spent his childhood at a royal court, and a
semi-barbarous one at that. This fact undoubtedly put its stamp on his per-
sonality and way of thought; he became in every respect an upholder of
authority and conservatism. At an early age he lost his father, and his
mother, Phasstias, retired to her native city and brought up her children there.
Aristotle received his earliest education, in accordance with the ancient As-
clepiad tradition, from his father's colleagues, who initiated him into the
biological and medical learning of their profession. It was necessary, how-
ever, for a properly educated physician to receive also philosophical instruc-
tion, and for this purpose Aristotle was sent at the age of eighteen to Plato's
Academy at Athens. There he remained for twenty years, was initiated into
the teachings of his master, wrote his first treatise, and already at that
period began to oppose his master's authority, which the latter is said to
have observed with no little displeasure. After Plato's death he was passed
over at the election of a successor as head of the Academy, in spite of his
already established reputation, and retired to Asia Minor, where he settled
down at the court of the Persian vassal-prince Hermeias of Atarneus, who
gave him his niece in marriage. Some years later, however, Hermeias was
deposed under a revolution, and Aristotle had to flee to the country of his
birth, Macedonia. There he was charged with the task of educating the heir
to the throne, Alexander, the future conqueror of the world, and he held this
post for three years (338-335). What influence the master exercised on the
pupil it is of course difficult to decide; the relations between them, however,
were on the whole good, though Alexander's increasingly despotic character
and barbaric outbursts of passion must have off'ended the cultured, self-
controlled Aristotle. His profession as teacher at any rate brought Aristotle
illustrious honours and made him a wealthy man, able to choose his place
of abode and his sphere of activity. He then moved back to Athens and lived
there under the protection of Alexander, highly esteemed by a constantly
34
CLASSICAL ANTIQUITY, MIDDLE AGES 35
increasing throng of pupils for the space of twelve years. During this period
he displayed indefatigable activity. He was granted the right to use for edu-
cational purposes a temple dedicated to Apollo Lycieus, after whom the place
was called the Lyceum, the archetype of learned educational institutions
throughout the world. Here every morning Aristotle gave scientific lectures
to his chosen pupils, often old and highly reputed men of science, who col-
laborated with him; and, further, every evening he held more popular courses
for younger collegiates. Moreover, he found time to write an incredible
amount on very different subjects: logic, metaphysics, art, politics, psychol-
ogy, and biology. This extraordinary activity testifies to his inexhaustible en-
ergy and splendid powers of organization. It is obvious that his disciples had
to carry out the rough work. Aristotle kept aloof from public life; indeed,
he was a foreigner in Athens. He was a conservative and a monarchist, how-
ever, and when after Alexander's death Athens rebelled against the Mace-
donian supremacy, his position became dangerous. For lack of other means
of calumniating him he was accused, like Socrates, of "godlessness." In
order, as he himself said, to save the Athenians from committing a fresh crime
against philosophy he fled to the island of Euboea; there he died shortly
afterwards, in the year 3x1. In external appearance he is said to have been of
small stature and corpulent; his carriage was proud, his manners arrogant and
sarcastic, his dress and way of living courtly, refined, and elegant. These
latter characteristics brought him personal enemies, who sought to blacken
his character. It is not possible, however, to bring any serious accusations
against him as a private person. It is true he appropriated with considerable
lack of bias the results of the work of earlier philosophers, but the ideas of
literary copyright were not so strict as they are now. On the other hand, he
treated different thinkers with true humanity; his polemics, when he went
in for them, were always courteous and his arguments founded on facts.
Towards his family, his friends and pupils, and even his slaves he was affec-
tionate and considerate.
Aristotle's sphere of activity was, as mentioned above, extraordinarily
extensive, and equally universal has been his influence during these thousands
of years in such widely separated spheres as biology, metaphysics, statesman-
ship, and art. In the present work it is of course possible to deal at any length
only with his biological work; besides this, however, we must touch upon
his general ideas so far as they affect his views on life in nature. And as varied
as his own interests have been the judgments passed on him by others; he
has been by turns elevated to the skies and dragged in the mud. On the whole
the biologists have been the most loyal; up to recent times, and not least in
our own day, he has had devoted admirers, and his works have been remark-
ably free from the bitter criticisms which biologists of more recent times
(Linnaeus, for example) have passed on his predecessors. The philosophers
36 THE HISTORY OF BIOLOGY
proper have judged him far more sternly; they have realized far more clearly
the deficiencies in his system of thought, and the weaknesses inherent in its
structure. The idealistic philosophers have accused him of vitiating Plato's
theory of ideas, while adherents of critical philosophy have criticized his
dogmatic ideas of the universe. His extraordinary influence both on his con-
temporaries and on posterity, however, no one can deny.
Form the true reality of matter
As a thinker Aristotle bases his system on Plato. According to Plato the ideas
of eternity are existing realities, of which the things of our earth are an im-
perfect image. Aristotle adopted the theory of ideas, but sought to overcome
the difficulty arising out of the questions how ideas are really related to
things and how they influence them, by placing ideas not outside things as
something independent and apart from their existence, but in things them-
selves. And he regarded the form of everything as its idea, as its true reality.
Form is the thing's reality, matter is a potentiality, to which form gives
reality. The bronze of which the statue is made is a potentiality; the form
which the sculptor gives it makes of the statue a reality. This method of
observation, derived from human life, Aristotle applies with inflexible con-
sistency to the whole of nature, both animate and inanimate. Thus the seed
is a potentiality out of which the germinating plant develops reality. The
same is true of the t^g and the embryo in relation to the creature which
develops therefrom. Consequently every lower stage of development is a
potentiality in relation to the higher stage of development which represents
its full realization. Thus we gtt a whole series of stages of development, be-
ginning with completely formless matter, which is an exclusive potentiality
without any reality at all, through inanimate nature, in which matter is
stronger than form, to the animate, in which form governs matter. Form in
living creatures is the soul, and the more highly developed it is, the more does
it control the corporeal matter. Plants have a lower kind of soul, which only
lives, but does not feel; animals possess a higher, sensitive soul, and, finally,
man has conscious reason. The means whereby form gives expression to its
dominance over matter is motion; all that occurs in the universe is motion,
and the more form-perfect the motion is, the higher the development it
represents. Our earth, therefore, with its manifold irregularities, is a lower
form of existence than the heavens, whose celestial bodies possess the most
perfect motion, circular motion. The heavenly bodies keep their position
and motion owing to their being enclosed in transparent spheres, one outside
the other. Thus they represent to the mind of Aristotle, as to that of Plato,
a higher form of existence than the earth with its creatures, including man,
and outside the outermost celestial sphere is the world of form free from all
matter, the highest intelligent existence, God, the fundamental origin of all
motion. Since, then, existence has its origin in a supreme intelligence, it is
CLASSICAL ANTIQUITY, MIDDLE AGES 37
natural that everything that happens has an intellectual cause, and every-
thing exists to serve a given purpose. This purpose is, above all, the develop-
ment of a higher form, the striving towards a higher intellectual existence.
Natural necessity and its cause, chance, have, it is true, their part to play on
the earth, but only as an attendant of incomplete matter; in the heavenly
spheres nothing happens by chance, but all is intellectual, while on the
earth the higher intelligence is victorious over its lower opponents. But con-
sequently terrestrial life, including man, is also governed by the higher
intelligence of the heavenly bodies, which get their impulses direct from
God.
The first evolutionist
Thus Aristotle makes his biological theories a link in the general cosmogony
which he built up on the fundamental principle of the domination of form —
that is, of the spirit over matter, and of motion as the origin of all things.
The whole world of this thought-structure is as foreign to our modern ideas
as it could possibly be; Democritus' atomic theory is far nearer our own
notions. Nevertheless, Aristotle's theory implies an absolute advance in the
sphere of biology. Here we find enunciated for the first time a really complete
theory of evolution. To the old natural philosophers, Democritus among
them, existence was a casual change of different forms. Again, Aristotle saw
a consistent evolution from lower to higher forms of being, and although it
is based on purely metaphysical speculation, this idea has proved for all time
a fertile one in the biological sphere, for the very reason that it is here in
agreement with actual fact. Quite in accordance with this his fundamental
principle, Aristotle also made special investigation into the development of
animals from the tgg and embryo to the perfect state, and in this sphere he
has made his most important contribution to biological research. But other-
wise his philosophical and educational activities embraced the whole of
biology, as it was known at that time, as well as all natural phenomena in
general. Of his purely biological works the following are extant: ten books
On the History of Animals, of which, however, three are considered spurious;
four books On the Parts of Animals; five books On the Reproduction of Animals;
and three books On the Soul. In these treatises he has collected all contempo-
rary knowledge of animal life, not only his own and his pupils' personal
observations, but also all the knowledge that his extensive collections of
books could impart regarding the observations of the earlier philosophers.
All this material he worked up with a view to including it in his general
cosmic theory. It has been said that "never before or since has a scheme been
so completely carried out with a view to incorporating biology in one com-
mon science, while at the same time by personal observations and literary
notes systematically building it up into one unit out of the phenomena"
(Burckhardt). This is true, and the reason for it is to be found in that unique
38 THE HISTORY OF BIOLOGY
genius for the purely formal side of thinking which belongs to Aristotle
alone; he is in fact the founder of formal logic and in that sphere his laws
hold good to this day. He is, therefore, also the originator of biological
classification, not only because he determined those categories in which
human thought has since primitive times sought to arrange natural objects,
and because he subordinated these to the general laws of thought which he
created, but also in that he sought to interpret and combine into a law-
bound whole those phenomena which accompany life in all its forms. By
this work he has paved the way, as no one else has done, for the further
development of biology as a science based on fixed principles. And his influ-
ence has extended to our own day; many of the biologists of our own time
have revived certain expressions out of his terminology and have been influ-
enced more or less directly by his ideas, particularly in regard to evolution.
But his great gift for form has also its darker side. He who saw in form the
true content of existence could not imagine a world to be other than finite —
spherical, for the sphere is the most perfect form — and as he could not visual-
ize an infinite world, he could not imagine infinite potentialities of knowl-
edge; on the contrary, he expressly declared that his own system, complete
as it was, would make it possible to solve all problems. But for that reason
he takes up all questions for discussion, even such as in our time would
be received with the old proverb "A fool can ask more than seven wise men
can answer." Often when reading his writings one can fancy that one hears
an inquisitive pupil offering objections which the master takes up with un-
disturbed calm and answers with unerring assurance in accordance with the
principles of his system; as, for instance, why men, and not women, be-
come bald; why the sow produces many pigs while the cow bears only one calf,
etc. The result is that, while the reader can sometimes trace in his writings
the pen of a biologist with almost a modern view of life's phenomena; on the
next page he may receive the impression of a master of scholastic disputation
from a mediaeval university. Many of the irregularities may certainly be due
to his treatises not having been carefully planned out; a number of them are
probably notes of lectures published by his pupils, and as such are perhaps
in places based on misapprehensions of the meaning of the lectures.
To give a complete description of Aristotle's biological works would be
a voluminous task and would result in a wearisome mass of detail. In order,
however, to give some idea of the peculiarities of his work, a brief account
must here be given of his position as regards the various main sections of
biology.
Aristotle's classification of animals
In regard to the classification of animals, Aristotle, as was hinted above,
has made an essentially important contribution to the subject by differentiat-
ing between and analysing and characterizing, from different points of view.
CLASSICAL ANTIQUITY, MIDDLE AGES 39
a number of systematic categories. "Animals may be characterized accord-
ing to their way of living, their actions, their habits, and their bodily parts. ' '
According to the three first-named principles they are divided into land-
animals and aquatic animals; certain of the latter live entirely i.'i the water —
the fishes; others live most of their time there, but breathe and breed outside
it — otters, beavers, crocodiles. The water-animals can partly swim, partly
creep, and in part they are adherent. The land-animals have similar charac-
teristics as regards habitat and way of living, feeding, habits, and character.
The most important bases of classification, however, are the parts of ani-
mals' bodies, both external and internal: motive organs, respiration, organs
of sense, blood-circulation. By combining various qualities the groups are
defined and characterized. These groups are variously extensive: "Many
animals allow of association into large divisions, such as birds, fishes, and
whales," and, further, ink-fish, shell-fish or mussels, and crayfi:li. Others
are more difficult to classify, such as the quadrupeds, which may certainly be
classified as oviparous and viviparous, but amongst these it is not possible
to make subdivisions, the animals having to be characterized each separately.
The categories which Aristotle thus established he never summarized; his
tabulated and generally recognized "system" has been extracted from his
writings by others and need not be repeated here, all the more so as it is
reproduced in different ways by different authors. Otherwise, his systematic
categories are only two in number, the genos and the eidos, the latter corre-
sponding to the individual animal form — horse, dog, lion — the former to
all combinations of a higher degree. That is really the reason why his sys-
tem cannot be compared with the Linnxan, with its manifold categories,
though it by no means detracts from its pioneer importance for all time.
His knoivledge of forms
In connexion with the system of Aristotle a few words may be said about
his knowledge of form — about the material out of which he built up his
system. In his writings have been recognized about 5x0 of the species which
present-day zoology has classified. These forms all belong to Greece and its
seas, and it seems that marine fauna interested him more almost than land
fauna; fishes, molluscs, and crustaceans are better represented in his works
than land-animals — in sharp contrast to what was afterwards the case with
Linnasus. Exotic animals Aristotle knows only from the descriptions of
others; it has often been stated that Alexander used to send him material for
investigation from the countries he conquered, but this can hardly be true.
The crocodile, for instance, he describes in the exact words of Herodotus
and he accepts as true without further comment the latter's statement that
the upper jaw of that creature is jointed on the lower jaw — which he would
never have said had he seen a crocodile. Still more remarkable is the fact
that he apparently never saw, or at any rate never carefully examined, a
40 THE HISTORY OF BIOLOGY
lion; in fact, he says of this animal that it has no cervical vertebras, but in-
stead has a single conjoining bone. In another place he declares of a lion's
bones that they are so hard that they give off sparks when struck, like flint,
and that they are "said" to have no medullary cavity. How much Aristotle
in general borrowed from the wisdom of his predecessors it is impossible to
determine, as their writings have been lost and he never quotes others except
with polemical intent. But examples such as those cited above undoubtedly
testify to a quite uncritical exploitation of foreign sources, and it is manifest
that the value of Aristotle lies not so much in the facts he established as in
the systematic working-up of the scientific material he had at his disposal.
And this systematizing work of his was of course as comprehensive as was
possible with the means available at the time. It is not only the outward
form of animals and their existence that interests him; he studies the migra-
tions of birds and the wanderings of fishes and tries to discover the causes
that underlie these habits; he critically examines the outward manifesta-
tions of animal intelligence, and everywhere he closely compares different
forms of life.
He chiefly occupies himself, however, with the anatomical and mor-
phological structure of animals, as well as with their reproduction and
evolution. These two spheres of investigation he has elaborated with the
utmost care, and here we find he stands out most prominently as the founder
of comparative natural philosophy. His t/eatises on human anatomy have
been lost, but his great descriptive work on animals is extant and deals
mainly with anatomical questions. Here he at once, in the very beginning of
the work, lays down that anatomical research should be comparative; the
less known should be studied by comparison with the better known, and
since the structure of the human body is best known, that should be the
point of departure. Following this method, he goes through the parts of the
body, the external as well as the internal. He bases his general anatomical
ideas on the same principles as Empedocles and the Hippocratic writings,
and certainly also as Democritus. With the first-mentioned he holds that all
beings are composed of four elements, while for him, too, the contraries hot
and cold are of fundamental importance. His description of human anatomy,
which forms the first book of his work on animal life, is of unequal value
and, as regards its details, undoubtedly very much dependent upon the state-
ments of others; but his manner of summarizing his description is excellent.
As to his description of the internal anatomy, he frankly acknowledges that
it is founded upon conclusions drawn from the dissection of animals, and,
broadly speaking, it is not very different from the Hippocratic writings re-
ferred to above. Thus the heart is to Aristotle the organ of the soul and the
intelligence, the brain serves the purpose of producing mucus and cooling
the blood; in this respect his ideas are inferior to those of both Democritus
CLASSICAL ANTIQUITY, MIDDLE AGES 41
ind Plato. The circulatory system is described in the same way as in Hippoc-
rates. The various parts of the digestive canal are described in some detail,
but as regards the physiology of the digestive process he has extremely
primitive and vague ideas, which is not to be wondered at, seeing that
the science of chemistry did not yet exist. "Cooking" plays the essential
part in his physiology. The food is "cooked" in the intestinal tube; the
heart pulsates through the regular "ebullition" of the blood. With regard to
the nervous system his notions are equally vague; as is indicated above, the
brain is cold, the spinal marrow is hot; nerves and tendons are confused. The
aural cochlea is described, as are also the membranes of the eye; the moisture
of the eye is believed to receive the visual impressions. It is curious to note
the amount of popular superstition that is accepted, as, for instance, predic-
tions read from the lines of the hand, or the idea that flat-footed people are
of a treacherous disposition. In the comparative anatomy and morphology
of animals Aristotle shows his many-sided interest in all kinds of life-forms
and his immense power of combining observations of various qualities with
striking characteristics. "Four-footed beasts which produce their young
alive have hair; four-footed beasts that lay eggs have scales." "A single-
hoofed animal with two horns I have never seen. . . . No animal has at the
same time tusks and horns." He also makes many sound observations re-
garding birds and reptiles, as, for example, in reference to the outer structure
of the sensory organs. A distinction is made between whales and fishes, and
the gills forming the breathing-apparatus of fish are described with emphasis
on the difference between osseans and sharks. Of the lower animals — the
bloodless, as Aristotle calls them, after the example of Democritus — the
ink-fish in particular is minutely described, many carefully observed details
being given. Crayfish, too, and insects are cleverly described in part, though
with some inaccurate details.
Reproduction of animals asexual and sexual
In his work on the reproduction of animals Aristotle differentiates between
animals which reproduce themselves by sexual means, by asexual means,
and by spontaneous generation. The latter occurs in a number of lower
animals which are produced out of putrefying substances; among these are
specially mentioned certain insects, such as fleas, mosquitoes, and day-flies
(other insects, such as grasshoppers, wasps, and flies, have sexual reproduc-
tion). Among the shell-fishes some are produced by asexual means through
bud-formation, others through self-generation. The possibility of this latter
method is explained by the fact that the whole of nature is full of life-spirit
or "soul"; this, under certain circumstances, gives form to the inanimate
matter and so gives rise to new beings. Sexual reproduction, again, is due to
the occurrence of male and female individuals. Of these two the male — or
more properly the man, for Aristotle takes as his starting-point here, as
41 THE HISTORY OF BIOLOGY
always, man — represents the more complete, "warmer" element, and the
female, the woman, the more incomplete, the "colder" element. The mascu-
line represents, above all, form, motion, activity; the feminine is matter,
the passive, and consequently potentiality, which achieves reality through
form. It is expressly asserted that the earth represents in the universe the
womanly and maternal, the sun the manly — • that is to say, the ancient
view held by most natural religions. The male sex-product, the seed, is a
product of the blood, which through complete "cooking" receives the
purest and most form-creating qualities. The woman's sex-product is the
menstrual blood, which is an undeveloped sperm — "half cooked," because
the woman is weaker, "colder" than man, and has not the power to cook her
product completely. In impregnation the man contributes to the future child
form, motion, soul; the woman matter, body; his contribution is compared
with the work of a carpenter, hers with the timber of which things are made.
Holding this view of reproduction, Aristotle, when making his thorough in-
vestigation into the question of heredity, can of course only involve himself
still deeper in abstract speculations. Against the opinion of earlier philoso-
phers that the seed is derived from all parts of the body and therefore gives
rise to similar individuals as issue, he asserts that on the contrary the seed
goes to the different parts of the body, through which process a remainder
is left over for the next generation, "as with a portrait-painter, a certain
amount of colour is left over similar to that which was used for the portrait. ' '
If the man's form-building power is sufficiently strong, the child will be a
boy; otherwise a girl; for that reason very young and very old fathers have
mostly girl-children. The cold north wind also favours the birth of girls, for
warmth is strength. The explanation of why children resemble partly their
parents and partly their ancestors is very complicated and ingenious, for all
its abstractness, but it would take too long to examine it here. Of far greater
value than these metaphysical speculations, at any rate, are Aristotle's
observations on the reproduction of animals. He draws up a scale in which
the animals are placed according to their development and points out that
those animals are highest which have a warm and moist, and not an earthy,
nature. For all animals with lungs are warmer than those without lungs,
and of those provided with lungs, again, the warmest are those which have
not tough, spongy, and anasmic, but soft and sanguineous lungs. The most
perfect animals, those which possess most warmth and moisture, whose
young are born alive and immediately start growing, are the mammals;
those which possess moisture, but less warmth lay eggs which afterwards
develop inside the female animal and are born alive: sharks. Warm and dry
animals lay "complete" eggs, such as birds and reptiles; cold and earthy ani-
mals lay "incomplete" eggs, such as osseans, frogs, and ink-fish; and finally
the lowest animals of all — that is, of those that propagate in a sexual
CLASSICAL ANTIQUITY, MIDDLE AGES 43
way — breed worms which give rise to eggs, as, for instance, insects, whose
pupas Aristotle regarded as eggs, whereas the true insect eggs were unknown
to him. His descriptions of animal development contain a mass of extraor-
dinarily sound details; as always, it is marine animals that interest him most.
His account of the breeding of sharks is especially well known, while the
pairing and growth of ink-fish are also described with a thorough knowledge
of his subject. Embryology, too, he discusses in great detail, chiefly the
development of the hen's egg, which Hippocrates had also studied; in particu-
lar the evolution of the heart and the first blood-vessels is carefully explained.
Various speculations on colour-variations, change of teeth, and other prob-
lems of development complete Aristotle's work on the reproduction of
animals, which, more than any other of his biological works, testifies to
both his greatness and his limitations.
H/s evolution theory really dogmatic
Aristotle's great contribution to the development of biological science
lies, as has already been pointed out, not so much in the sphere of discoveries
as in the thought-system, embracing all the phenomena of life, which he
created and consistently worked out in all its details. The finest merit of this
system of thought lies in the fact of its being based on an evolution subject
to rigid laws and proceeding from the lower to the higher. But as this theory
of evolution is, as has been shown above, primarily based on a predominant
guiding intelligence, it acquires a dogmatic arbitrariness; the subjection to
law is not an act of nature itself, but rather a product of divine wisdom, or,
in other words, human speculation. This could, then, it is true, solve with
abstract catchwords all the problems against which Democritus' atomic
theory was powerless, but any such method of solution failed to stimulate
thought to continued search; on the contrary, it induced a feeling of self-
complacent satisfaction with the limited cosmogony produced out of unreal
systems of thought. Thus it came about that Aristotle, the founder of sys-
tematic biology, became at the same time the father of the scholastic philoso-
phy of the Middle Ages; that the man who was the first to introduce and
logically to apply to the conception of the entire universe a theory of evolu-
tion from the lower to the higher appeared fifteen centuries later as the
founder of a system of stagnation and obedience to authority. What the
whole of this long period lacked was a conception of nature which would
have associated Aristotle's theory of subjection to law with Democritus'
theory of the dominance of necessity in nature. Subjection to law caused by a
personal, guiding will, and necessity with pure chance as its driving force:
the choice lay between these two alternatives until, through Galileo and
Newton, the impersonal, law-bound force, operating by natural necessity,
was made the basis on which to interpret the course of events in the universe.
For the fact that such a long time should have elapsed before this conception
44 THE HISTORY OF BIOLOGY
of nature as held today could make itself realized, Aristotle and the system
which he created, unexcelled in its perfection of form as it is, are mainly-
responsible. Under such circumstances it is not to be wondered at that the
biology of antiquity, in spite of its splendid achievement, never succeeded
in advancing beyond Aristotle's conception of the phenomena of life.
But if Aristotle thus represents, in most fields of science, the highest
that the culture of antiquity could attain, beyond which no further develop-
ment took place, this is not merely due to the influence of his own personality
and system. At the time of his death the Greeks had already seen their best
days. The civic spirit which, in spite of sordid party strife and sanguinary
border-feuds, had borne forward the petty Greek states both before and in
the throes of the Persian wars, disappeared as soon as the states lost their
independence, first through the hegemonies which the Athenians and the
Spartans in turn exercised over them, and later on through the Macedonian
and the Roman conquests. And with the sense of patriotism disappeared also
the intellectual power and will to act. The semi-oriental monarchies into
which the empire of Alexander became split up were certainly often governed
by enlightened princes who generously patronized the sciences, but their
lavish pensions had nevertheless to be purchased with obsequious flattery,
and the proud self-respect which induced Empedocles to refuse a royal crown,
and Heracleitus to decline the office of high-priest, existed no longer. The
great systems of thought which were created by the noblest spirits of later
antiquity were in fact essentially founded upon ethical aims; they were
intended to reinforce the individual in the struggle against the ever-increas-
ing difficulties which life in those days presented. The exact sciences, again,
were divided up more and more into special spheres and the research work
carried out during the succeeding centuries gave substantial results, until
here, too, the spiritual weariness from which that epoch suffered claimed
its due.
In these circumstances it may seem suitable in the following chapter to
pay special regard to the attempts at a general explanation of natural phenom-
ena which were made after Aristotle, after which we shall view the results
achieved by the biology of antiquity as a special line of research.
CHAPTER VI
NATURAL-PHILOSOPHICAL SYSTEMS AFTER ARISTOTLE
Aristotle' s folloivers
WHEN Aristotle fled from Athens he left his school in the hands of
Theophrastus, who had been his faithful friend and follower ever
since his student days with Plato. Though ten years younger than
his master, Theophrastus was an old man when he assumed the leadership of
the Lyceum, but he lived much longer and for more than thirty years presided
with honour over the education of the pupils. Already under Aristotle he had
paid special attention to the study of botany and he continued to work in
this science in the spirit of his master. His two treatises on plants are to
botany what Aristotle's works were to zoology. Furthermore, there is extant
a "history of physics" by him, which has always been the main source of
our knowledge of the ideas of the ancient natural philosophers. He also
wrote a zoological work, which has been lost, but on the whole it seems to
have contained nothing essentially new that is not found in Aristotle.
On the other hand, Theophrastus' successor, Strato, developed Aris-
totle's theory along entirely fresh lines. Unfortunately the present age is
acquainted with his point of view only through the references of other
authors, but from these it is clear that he was a truly independent thinker.
Born at Lampsacus in Asia Minor, he became a disciple of Theophrastus at
an early age, and after the latter's death held his professorship for eighteen
years. His numerous writings, now lost, dealt particularly with problems of
natural science and procured him the title of "the physicist." In contrast to
Aristotle he denied the existence of a dominant intelligence outside the uni-
verse; he imagined that the forces that govern the course of events dwell in
things themselves and operate by natural necessity. Further, the soul of
man he believed to be a force inhabiting the body, expressing itself as motion,
and having the brain for its organ. On the other hand, he seems to have at-
tacked Democritus' theory of the atoms and infinite space and considered
the whole world to be finite.
Strato's successors appear to have been men of little importance, and
although Aristotle's school survived down to the sixth century after Christ,
it nevertheless ceased to act as a guiding light in science; its teachers and
pupils became for the most part involved in specialized investigations into
grammar, literature, and ethics, and the keen interest in the natural sciences
45
46 THE HISTORY OF BIOLOGY
which the founder of the school had evoked disappeared entirely from their
circle. Instead, philosophical speculations on nature were undertaken by
another school of thinkers, who revived the atomic theory of Democritus,
which was thus given a fresh lease of life and survived not only the classical
period, but through the Middle Ages until the Renaissance. This new line of
thought was directed by Epicurus the Athenian, one of the most discussed
philosophical personalities of antiquity. He lived between the years 34x and
zji B.C., and in his native town founded a school whose members lived
quietly and carried out joint researches under their master's guidance. As
already mentioned, his theoretical standpoint was Democritus' atomic
theory, which he adopted without really developing it further. In direct
opposition to Aristotle he taught that universal space is infinite, that bodies
are composed of particles indivisible in themselves, whose motions are the
cause of everything that happens and through whose alternate association
and dissolution worlds arise and perish. Even the soul of man consists of
atoms and is thus a purely corporeal organ. There is no universal intelli-
gence, but all things happen through natural causes. What these causes are
Epicurus did not bother much about. In fact, he considered it hardly worth
while trying to find out the secrets of nature; thus it might well be, he ex-
pressly assures us, that the moon borrows light from the sun, but it might
equally well be self-illuminating. The main thing was that one assume a
natural explanation of the world and the universe; it mattered little what
this explanation turned out to be like in detail if only man rid himself of
the superstition which always accompanies a belief in supernatural powers.
Epicurus' system, in fact, was expressly based on the idea of creating by the
aid of philosophy a pleasant existence, and man attained this best by being
an opportunist in all the main problems of life. That on account of this he
gained numerous followers is generally acknowledged, but also the principle
of a cosmic conception which lays most stress on the exclusion of all super-
natural elements, and in doing so does not bother much about the difficulties
arising out of questions of detail, has always had, and will indeed always
have, keen adherents. It is no matter for surprise, however, that the follow-
ers of Epicurus, with such a view of the functions of scientific research, could
never claim any really great natural philosopher of antiquity amongst their
number. Epicureanism survived as a mode of living which invited people
during depressing and hopeless times to seek life's happiness and goal in
pleasure, spiritual as well as material. To Epicurus himself and to his friends
pleasure was essentially spiritual; their material needs were extremely
modest. But matters became worse when his doctrine was brought to Rome.
In that world-capital it degenerated into an unbridled worship of pleasure,
particularly under the Empire; the fact that Nero and his friends called
themselves Epicureans was not calculated to heighten the school's reputa-
CLASSICAL ANTIQUITY, MIDDLE AGES 47
tion. Even in Rome, however, it gained adherents of a nobler character. One
of these, Lucretius, deserves further mention, especially as his enunciation of
the atomic theory is the most detailed of its kind that has been handed down
to us from antiquity, and as such it is also of interest on the grounds of the
biological particulars which it contains.
Titus Lucretius Carus was probably born in 99 and died in 55 B.C. He
belonged to a famous patrician family and seems to have been acquainted
with several well-known persons among his contemporaries, but although
the epoch in which he lived — he was contemporary with Cassar and Cicero —
is without doubt the best known in antiquity from a historical point of
view, nothing is known of his life with any certainty. He seems to have kept
aloof from the political struggles of his time and devoted himself entirely
to philosophical and literary study. An early father of the Christian Church
declares that he died by his own hand; the statement may indeed be true,
for in the deeply unhappy age in which he lived, this desperate way out of
life was, as is well known, resorted to by many. It was not until after his
death that his great work On the Nature of Things was published, in which he
recorded the results of his philosophical speculations. Following the example
of the earlier Greek philosophers, particularly of Empedocles, whom he
greatly admired, he has clothed his thoughts in verse form; he is the last of
the ancient philosophers to do so. The wealth of imagination and the high
inspiration which fill his poem have given him a place amongst the greatest
poets of antiquity, but he is of great interest also as a philosopher. The most
striking feature of his poem is his passionate love of truth and his absolute
conviction that thought ultimately will succeed in penetrating the true
nature of things. To the glorious mission of philosophy to seek after truth
he opposes the grim picture of darkness and superstition called forth by
traditional theism, a miserable state from which he hopes free-thought will
save humanity. The cosmic explanation which he accepts as the only pos-
sible right one is the atomic theory, in the form in which Epicurus expounded
it; Democritus is mentioned only in passing. It can hardly be said that he
contributed anything towards the development of the general principle of
that theory, but at any rate his conception differs from that of which an
account has been given above in connexion with his predecessors. The worlds
are infinite in number, formed of atoms moving in empty space. Their mo-
tion is due to gravity and consequently represents a constant descent; the
fact, however, that they strike against one another, as presupposed in the
atomic theory, is due to their fall's being for internal reasons not quite
perpendicular, but deflecting to one side.
Lucretius' soul theory
The most interesting of all, however, is Lucretius' attempt to apply the
atomic theory in detail to the phenomena of the senses and the processes of
48 THE HISTORY OF BIOLOGY
the soul. Following Aristotle, with whose theory of the finality of all
things he nevertheless sharply disagrees, he assumes three kinds of soul,
animus, mens, and anima'^ — that is, the spirit, the understanding, and the
soul or life-principle. He does not, however, consistently differentiate be-
tween these three categories, but discusses them mostly as one single idea.
In quality the soul is material, an organ, like the rest, formed of extremely
small atoms distributed throughout the body, very mobile, and therefore
easily dispersed. These atoms are of three determinable kinds: warmth, air,
and "aura," a more rarified kind of air corresponding to the "pneuma" of
Hippocrates. But besides the three categories of atoms just named, the soul
contains still a fourth component, which has no name, but which forms the
real percipient, the consciousness in the soul, and whose atoms are the small-
est and most mobile. They give impulse to the other soul-atoms and thereby
indirectly to the movements of the body. These component parts of the soul,
being variously commingled, produce the varying soul-characteristics in
different individuals and make themselves felt in different degrees in the same
individual in different states of mind; heat in anger, cold air in terror, etc.
The soul, which in life is contained by the body, as a vessel contains whatever
is kept therein, dissolves at death into the simplest component parts and is
annihilated. The immortality of the soul as maintained by Plato and his
disciples is attacked by Lucretius with passionate intensity. Again and again
he seeks to prove that this, combined with a belief in gods, is the cause of all
human miseries. Sense-perceptions are, according to Lucretius, due to things'
giving off from their surface a kind of light particles which, formed like the
things themselves, float about in space and influence the organs of sense. As
a proof that such images are given off he cites, inter alia, the change of skin
in snakes and insects. All sensations — sight, hearing, smell, and taste —
are thus excited by different atoms which affect the organs of the body. The
ideas arising herefrom are caused by a mass of still more subtle images of
things, floating about in space even after the things themselves have dis-
appeared. Thus one sees in imagination images of individuals long since
dead, and owing to the images' sometimes coalescing, one receives impres-
sions of creatures which have never existed in reality; for instance, through
the coalescing of a horse and a human image is produced the mythical
centaur. Through such images still remaining in the soul dreams arise. Primi-
tive as these sensory physiological speculations are, they are nevertheless
accompanied by a number of extremely striking observations regarding differ-
ent kinds of sensations. In particular Lucretius discussed in detail sensations
in the sphere of sexual life, which he describes with a curious mixture of
^ These names, which Lucretius undoubtedly himself invented on the Greek model, are
again found in the psychological terminology of the Middle Ages, and even in Swedenborg.
CLASSICAL ANTIQUITY, MIDDLE AGES 49
minute observations from a natural-scientific point of view and poetic
inspiration.
Lucretius' influence upon posterity has been both lasting and important,
It is undoubtedly due mostly to him that atomism survived throughout the
Middle Ages, although in obscurity, owing to the hostility of the Church.''
During the Renaissance he was held in high estimation; the greatest thinker
of that epoch, Giordano Bruno, was strongly influenced by him and in imita-
tion of him wrote several of his scientific works in verse form, and even the
free-thinkers of the eighteenth century studied him closely. Yet it can hardly
be said that he advanced the natural sciences. He has not succeeded in improv-
ing upon the atomic theory as created by Democritus; such progress as Aris-
totle made in the sphere of biological development he rejected with the
theory of finality upon which it rested, but without succeeding in substitut-
ing any better subjection to law. On the whole, atomism became a theory that
brought no benefit to natural research until in the beginning of the nineteenth
century it was, through Dalton, adopted in chemical research and, thanks
chiefly to Berzelius, became the most fruitful working hypothesis of that
science. Since then it has been one of the most important foundations on
which our idea of nature, both inorganic and organic, is based ; but the univer-
sal application which the ancient atomists ascribe to it it has not received;
no true natural philosopher of today hopes to be able to explain the phe-
nomena of animate life with its aid, although in quasi-scientific popular
literature attempts have been made to do so.
^ In his Divina Commedia Dante relates that in hell there were several thousand "Epicu-
reans," among them many of the most eminent men of his own time.
CHAPTER VII
SPECIALIZED BIOLOGICAL RESEARCH AFTER ARISTOTLE
The anatomists of Alexandria
IF, THEjsT, the ancient conception of nature failed to advance beyond the
point to which Aristotle brought it, nevertheless there developed after
his time and on the foundations laid by him a specialized form of biologi-
cal research which during the following centuries produced rare and abun-
dant harvests. The centre of natural research during this period, not only in
the biological, but in other branches, was Alexandria, the purely Greek
capital of Egypt. Under the patronage of the refined and generous kings of
the Ptolemaic dynasty there was here established an institute of scientific
research the like of which the ancient world never saw before or afterwards.
Even the founder of the dynasty, Ptolemy I (died in 283 b.c), was a highly
cultured man who collected books and was himself an author. His son Ptol-
emy II founded the Museum of Alexandria (mouseion — a temple of the god-
desses of song and wisdom, the Muses), an institution where scholars from
every country received lodging and maintenance and substantial assistance
for the furthering of their research work. It was conducted on the lines of
an academy with the chief librarian as chairman; the highest authority was
exercised by the high-priest of the Muses, who was religious head of the
college. All the branches of science known to classical antiquity were studied
here; the science of biology was chiefly pursued in connexion with medicine,
like anatomy and physiology.
It has been mentioned above that medical science was from early times
highly developed in Egypt, inasmuch as the custom of embalming bodies in
that country contributed towards increasing the knowledge of human anat-
omy. Herein, then, lay certain preconditions for the stimulus given to the
study of anatomy in Alexandria; substantial grants of money and literary
and material aid from the princes rapidly advanced the development of the
school of medicine. Two teachers of more than usually prominent gifts,
Herophilus and Erasistratus, finally brought to the school a reputation such
as no other attained in classical times. We know little of the personal history
of these men, and not much about the general ideas of nature which they
embraced. It is assumed, however, that they were influenced by the scepticism
of Pyrrho, who apparently found most of his adherents amongst the Alexan-
drian physicians. Pyrrho of Elis (376-188 b.c.) taught that no knowledge
50
CLASSICAL ANTIQUITY, MIDDLE AGES 5I
of things is really possible; man can know nothing and prove nothing, not
even the impossibility of knowledge or justification for doubt. Such a funda-
mental principle naturally precluded any theoretical conception of nature,
whether Democritean or Aristotelean, but it was just this very circumstance
which drove a philosopher seeking after knowledge to become all the more
deeply engrossed in specialized science and its practical application. It was
thus exclusively through detailed anatomical research that the Alexandrian
medical school advanced the science of biology.
Herophilus was a native of Chalcedon in Asia Minor, studied in the
Asclepiad schools in Cos and Cnidus, and afterwards worked as a teacher and
researcher in Alexandria. The dates of his birth and death are unknown, but
his activities fall within the decades about the year 300 b.c. His writings,
too, are lost, except for a few fragments; their contents are known to us only
through the references of other authors. That Herophilus was one of the
most prominent anatomists of antiquity is, however, universally acknowl-
edged, both by classical and by modern authors. His fame is based on the
numerous discoveries he made, particularly in human anatomy. Every part
of the human body was investigated by him, and what more than anything
else attracted the attention of his contemporaries was the fact that he em-
ployed human bodies for the purposes of investigation. Sceptics as they were,
he and his pupils despised the traditional fear of dissecting human bodies,
and the enlightened Ptolemaic rulers placed material at their disposal. It is
even declared that Herophilus took advantage of opportunities offered to
him to carry out investigations on living human beings — criminals con-
demned to death, whose internal organs he studied in a living state. ^ Among
the organs which he described in detail may specially be mentioned the brain;
he discovered and gave an account of its membranes and its venous blood
sinuses, which still bear his name, the torcular Herophili. Moreover, he
studied the ventricles of the brain, being particularly interested in the
fourth, which he regarded as the organ of the soul. He gave similar close
study to the eye, its membranes, film, and retina. He also described the ali-
mentary canal; the name "doudenum" for its upper section comes from him.
The liver he carefully studied with regard to the variations in its shape in
different individuals. The circulatory system he also made the subject of close
investigation; he compared the walls of the arteries and the veins and studied
the pulse at different ages and under different bodily conditions. That the
arteries contained the pneuma he believed in common with all other re-
searchers of his time. For the first time he cleared up the question of the dif-
ference between nerves and tendons. Finally, he carefully worked out the
' One of the early Fathers, Tertullian, quotes, among other heinous acts committed by
the heathen, that Herophilus tortured to death six hundred persons — a story on a par with
much that is related nowadays in anti-vivisectionist literature.
5X THE HISTORY OF BIOLOGY
anatomy of the genital organs. His physiological ideas were governed by the
usual conceptions of antiquity — four different life-elements localized in
corresponding main organs; they are therefore of no special interest. For the
rest, Herophilus was a great admirer of Hippocrates, whose views on dis-
eases and remedies he accepted without reserve.
Contemporary and in competition with Herophilus was Erasistratus of
Cheos, a small island in the ^gean. His dates and personal history are as
unknown to us as those of Herophilus, and his writings were lost even in
late antiquity. According to a late and unconfirmed report he was the nephew
of Aristotle; it is certain that his teacher, Metrodorus, was the latter's con-
temporary and friend. Erasistratus began his career as court physician to the
Seleucides of Syria, but he was called thence to Alexandria, where he founded
a school of medicine. His anatomical works dealt chiefly with the circulatory
system. The heart he studied with care and gave its valves the names they
still bear. Further, he established the connexion between arteries and veins
and explained the bleeding from the arteries in wounds by the assumption
that their pneuma disappears, and in its stead the blood from the venous
system penetrates into the arteries and then flows from the wound. Again,
he investigated the lymphatic ducts and the secretion of chyle in live ani-
mals. He made important discoveries which increased the knowledge of the
nervous system; he distinguished between the motor and sensory nerves and
was the first to describe in detail the convolutions of the brain. As a physician
Erasistratus was more practical than Herophilus; he utterly scorned the Hip-
pocratic traditions, prescribed simple remedies, avoided venesection, and
strongly advocated a hygienic mode of life.
Hostility bettveen the medical schools of Alexandria
This opposition between the two Alexandrian anatomists had fateful con-
sequences for science. They themselves impugned one another by polemics
and intrigue, while there existed a still greater hostility between the respec-
tive schools. The "Herophilites" drove their master's conservatism and re-
spect for Hippocrates to extreme limits, while the "Erasistratites" held up
to scorn and counteracted the virtues of the medical tradition. This was natu-
rally bound to prejudice science, and it was all the more disastrous as the
cultural conditions in Alexandria became in time seriously impaired. The
early enlightened Ptolemaic kings were succeeded by a line of degenerate
scoundrels who neglected the interests of learning as they neglected all their
other duties. The Museum declined, its grants were reduced, and the learned
often fell victims to the tyrants' whims. Thus finally Alexandria became a
provincial town within the great Roman Empire. The Museum certainly
survived, but without the encouragement which the native rulers had given
to it; it was eventually destroyed in a riot — the Alexandrine mob was known
as the most unruly in the whole of the Roman Empire — and in the end the
CLASSICAL ANTIQUITY, MIDDLE AGES 53
extremely fanatical Christian Church in Egypt wiped out the last vestiges
of pagan scholarship.
Roman natural science
In Rome, which in time assumed Alexandria's position as the supreme capital
of the world, there arose no equivalent to the Museum. It was not until a
later epoch that the Roman people, with their decidedly practical mind,
attained to the higher culture, and only in the juridical sphere did they make
any independent contribution to the development of intellectual work;
otherwise they appropriated Greek culture, special branches of which they
converted to various — mostly practical — purposes. One applied science
of this kind which the Romans created was the science of agriculture. In
contrast to the Greeks they were agriculturists body and soul and early felt
the need of having their experiences in this sphere collated and recorded.
The old censor Cato actually wrote a treatise on agriculture, and after him
there are mentioned a large number of writers on agricultural subjects. The
foremost of these was undoubtedly Columella, whose writings contain suf-
ficient of interest to biology to warrant his being mentioned.
Lucius Junius Moderatus Columella was born at the beginning of the
Christian era in Spain, but he seems to have lived in Rome and there com-
posed his treatise on agriculture in twelve books. Of biological interest is
his account of domestic animals, their management and necessities of life,
their races and areas of distribution. All the useful animals of his time, even
the bee, are dealt with in his work, most sections of which are, as a matter
of fact, of purely economic interest.
Pliny, too, shows a marked interest In the practical application of
science. He was the most eminent of Rome's natural philosophers and, next
to Aristotle, the most influential of the biologists of classical antiquity.
Throughout later antiquity and the Middle Ages and far on into more recent
times his Natural History has played an important part in the development of
science; indeed, it may be said that even in our own day his influence has not
entirely waned. In contrast to Aristotle, however, he has been harshly criti-
cized in the biological literature of the present day — extravagantly so, be-
cause more has been demanded of him than he ever intended and more than he
was able to offer. He has been characterized as a soulless compiler, because,
more honest than Aristotle, he always quotes his sources; his superstition has
been ridiculed because he tells of marvellous animals the existence of which
none of his contemporaries doubted. Above all, the constantly repeated com-
parisons between him and Aristotle are entirely unjustified; the aims and
methods of the one were not those of the other. A study of Pliny's life and
work will confirm this.
Gaius Plinius Secundus was bora a.d. i3 at Comum, now the Como
of northern Italy. He belonged to a family of public officials, and his own
54 THE HISTORY OF BIOLOGY
career was in fact an official career after the typically Roman model. He
received a thorough education from good private teachers in Rome and after-
wards served alternately in the army and the civil administration. He spent
a long time in Germany, where he held a military command, and eventually
became commander-in-chief of a division of the fleet. On the first occasion
known to history when Vesuvius was in eruption (a.d. 79), he lay with his
vessels near Naples. In order to study the unusual phenomenon closer he
ordered a boat to row him to the foot of the volcano, and there he perished.
Both in his treatises and in his biographies he stands out as a man of the
highest probity of the good old Roman type, brave, honest, and loyal. Con-
stantly engaged in a round of official functions, he devoted every unoccupied
moment, both at Rome and in the distant provinces, to study and author-
ship. His capacity for work is described as inexhaustible, and the writings
he left bear witness to his remarkable erudition. He worked at the most
widely different subjects: military science and military history, rhetoric, and
linguistics. The only treatise of his which has come down to posterity is his
great work of thirty-seven books, the Nattiral History^ on which his fame
really rests. It represents a veritable encyclopaedia covering the entire knowl-
edge of nature at that time, including its application to medicine, technology,
and economy. It begins with an account of the universe and its laws and goes
on to give an increasingly specialized description of various natural objects.
Books VIII to XI deal with animals, and a number of scattered notes on
zoological subjects are also to be found in other sections of the great
work.
In his general conception Pliny was a Stoic. The Stoic philosophy was
founded in Athens at the same time as the Epicurean, but its founders and
earliest leaders all came from Semitic countries, and several historians have
made an attempt to trace an oriental influence in its ascetic contempt for
material life-values and its strong feeling for personal responsibility. How-
ever, Stoicism found its way to Rome, whose noblest men were attracted by
its austere sense of duty and, as was the Roman habit, converted its system
to practical uses. Stoicism laid greater stress than Epicureanism on a practi-
cal way of living; it was not so much concerned with the general conception
of nature and its laws. Even Pliny's general ideas of nature constitute a not
very interesting record of the dicta of earlier authors; here, for instance, we
find the Aristotelean theory of a spherical universe, with the four elements as
its essential components, with — on the Pythagorean, or rather Heracleitean,
pattern — fire as the primary cause, the origin of the soul; a divinity governs
the world, but it is folly to seek to discover its entity, though a greater folly
still is polytheism. Oracular utterances and prodigies, on the other hand,
are recounted by the score and without any expression of doubt of their
value.
CLASSICAL ANTIQUITY, MIDDLE AGES 55
Pliny' s description of animals
The zoological section of his natural history is as encyclopaedically treated
as the rest. The various animals are enumerated without any sequence, but
on the whole the largest and most remarkable are mentioned first; they
are described with reference to their habits, their utility, and the mischief
they do, the date of their first being exhibited and employed in Rome, and
in general their relation to man. On the other hand, no attempt whatever is
made to give a true description of their external and internal structure. The
earlier biologists of antiquity, including Aristotle, were, as we know, not
very conspicuous for any important criticisms on points of detail, especially
where exotic animals are concerned, and Pliny with great goodwill collects
all the marvels he can find in earlier writings and narrates them without re-
serve. Consequently his account teems with the most fantastic fables. As an
example of the assertions he makes may be mentioned his description of the
elephant, which, characteristically enough, is named first of all animals.
"Amongst land-animals the elephant is the largest and the one whose intelli-
gence comes nearest that of man, for he understands the language of his
country, obeys commands, has a memory for training, takes delight in love
and honour, and also possesses a rare thing even amongst men — honesty,
self-control, and a sense of justice; he also worships stars and venerates the sun
and moon. In the mountains of Mauretania it is said that herds of elephants
move at new moon down to a river by the name of Amilo, ceremoniously
cleanse themselves there by spraying one another with water, and after hav-
ing thus paid their respects to the heavenly light return to the forests bearing
their weary calves with them. It is also said that when they are to be trans-
ported overseas, they refuse to go on board until the master of the ship has
given them a promise under oath to convey them home again." Further, they
are so modest that they never mate except in very secluded spots, while
adultery never occurs amongst them. Towards weaker animals they show
compassion, so that an elephant when passing through a flock of sheep will
with his trunk lift out of the way those he meets, for fear of trampling on
them.^ Besides these and similar childish statements accounts are given of
the habits of elephants and the way to tame them, which are quite correct,
as well as a number of facts of interest from the point of view of cultural
history as to their employment amongst different peoples, when they were
first exhibited at Rome, etc. Pliny likewise relates many wonderful stories
about other lesser-known animals, such as the elk and the aurochs of north-
ern Europe. On the other hand, the information he gives regarding the or-
dinary domestic animals of his own country is on the whole reliable and the
^ It was probably this description that directly or indirectly induced a Danish king in
the Middle Ages to found an Order of the Elephant, with the purpose of thereby exhorting its
members to imitate the admirable qualities of that noble animal.
56 THE HISTORY OF BIOLOGY
particulars of cattle-management at that period are quite sound. Among
land-animals he includes even the cold-blooded vertebrates; after that he
deals with birds, fishes (including molluscs), and insects. Details of all kinds
of animals belonging to these groups are in the main similar to those men-
tioned above. Insects in particular seem to have attracted his attention; he
never wearies of expressing his admiration for their perfect organisms, in
spite of their small bodily size, and he gives an account of what was known
of their organic systems. The bee he describes in great detail and he relates
its habits, in many respects correctly, although he did not succeed, any
more than the other ancient authors, in gaining any idea of its method of
reproduction.
Pliny' s anatomical ideas
After Pliny has thus given an account of the animals known to him, he dis-
cusses the various organs of the human and animal body on the same plan;
each organ is considered in reference to its qualities and occurrence in the
various animals. Here we clearly find Aristotle to be the pattern and main
source of information; but while the latter' s description of the organs is given
with a view to tracing the connexion and origin of the forms, Pliny's ac-
count still has the character of a work of reference, in which all the memo-
randa that he was able to collect out of his vast erudition are cited with no
kind of theoretical purpose and without any deeper significance than the
word that clothes them; for instance, the horn of the ox, of the horned snake,
and of the snail are treated as all one. A wealth of valuable notes from the
rich scientific store of the anatomical knowledge of antiquity has thus been
preserved for posterity by Pliny, whose sources of information have been
lost to us, but besides this he conscientiously notes down a number of ancient
prodigies, which of course inspired fear in his time and have consequently
been handed down to history — of sacrificial animals which had no liver,
or of the Messenian champion of liberty, Aristomenes, whose heart the Spar-
tans found to be covered with hair — without in any way hinting at the
possibility of fraud on the part of the sacrificial priests. A mass of information
regarding the medicinal use of animals or animal parts closes the zoological
section of his Natural History.
Pliny himself states that he had recourse to two thousand books by
various authors for the compiling of his work, which maintains also from
beginning to end its character of a confused motley of notes. This many-sided
learning, which has very much impressed past generations, seems in our day,
when only first-hand knowledge is really respected, rather pitiful. Never-
theless, as previously pointed out, Pliny has certainly been underestimated.
For fifteen hundred years his work was the main source of man's knowledge
of natural history, and when during the Renaissance a Gesner or an Aldro-
vandi revived the pursuit of zoological research, they at once began where
CLASSICAL ANTIQUITY, MIDDLE AGES 57
Pliny had left off and carried on the work after his method. In this way
present-day zoology, as regards the study of fauna and classification, takes
Pliny as its starting-point, just as in the matter of comparative anatomy and
morphology it is based on Aristotle, and therefore the services of the one
should in all fairness be recognized as much as those of the other, even if
they refer to entirely different fields of study.
CHAPTER VIII
THE DECLINE OF SCIENCE IN LATE ANTIQUITY
The decline of ancient civilisation: its causes
IT HAS ALREADY been pointed out that the natural science of antiquity
reached its zenith in Aristotle, and a number of reasons have been given
for the fact that only in points of detail, but never in regard to the sum-
marizing of the results achieved, did it advance beyond his standpoint. While
Rome, first as a republic and then as an empire, was conquering and adminis-
tering the whole of the civilized world, there began an era which, more than
any other, should have been devoted to promoting the work of intellectual
culture. The universal peace that prevailed during the first two centuries of
the Christian era has never had its counterpart either before or since, for the
border feuds and insurrections which disturbed it were entirely local and
transient. And as there was peace, there was also prosperity; even up to the
present day the ruins of buildings bear witness to the common and private
wealth of those days throughout the length and breadth of the Roman
Empire. And yet it was this very epoch which witnessed the decline of an-
cient science — indeed the whole of the culture of antiquity. It was not long
before the best minds in the intellectual world of the time realized this fact.
Pliny, for instance, is never tired of repeating that humanity is corrupt and
that his age was worse than the era that had passed. The reason he gives is
the increasing corruption of morals — an assertion with which innumerable
other ancient authors are in agreement and which has therefore been re-
peated in more recent times. The cause cannot lie there, however; moral cor-
ruption is always a symptom and not a cause of cultural decadence. The cause
is far more likely to be found in the change in the common conception of life
which was a consequence of subjection under the Empire. The ancient provin-
cial patriotism had lost its power to survive and there was no possibility of
any fresh form of social community developing; instead the individual per-
sonality appears as struggling for freedom from external oppression and
grievances. This self-assertion against an oppressive existence both Epicu-
reans and Stoics sought to put into practice, each in their own way; but, as
we have seen, their teachings formed no good soil in which to cultivate
empirical research. In the long run, however, the purely negative insensibility
to suffering which constituted the philosophy of life of these schools of
thought could not suffice; in their place appeared lines of thought start-
58
CLASSICAL ANTIQUITY, MIDDLE AGES 59
ing out from the idea of leading the personality into an existence differ-
ent from the earthly, of creating with the aid of some kind of higher,
secret knowledge a happier world for the soul to live in. There thus arose a
half-mystical, half-experimental psychology, which was nurtured by philo-
sophical schools possessing sectarian organizations, like that of the Pythag
oreans in the old days. One of these schools, and the most fantastic of all,
actually called themselves neo-Pythagoreans; another, more scientifically
serious, was the neo-Platonic, which sought to bring the human spirit, along
the mystical path of introspection, into contact with the world of ideas,
which Plato declared to be the only true world. Through this development
the very idea of philosophy became radically altered; the philosopher was no
longer a lover of wisdom, as the name implies, but a lover of piety. But as
such he retained no interest in natural phenomena; his spirit in fact lived in
supernatural regions of space, and if he devoted any time to the objects of
nature, it was merely in order to discover the secret divine powers which,
hidden from the eyes of the ignorant, dwelt in plants and animals.
For the belief in God awakened to new life during later antiquity; not
the old sacrificial faith ^ so indissolubly bound up with the inner life of the
petty states, but faith in a supreme power able to save the individual from
sorrow and suffering. Numerous religious brotherhoods were founded which
sought by mystical means to procure for their members peace and happiness
in this life, or at any rate in the life to come. Among these faiths appeared
Christianity, which was finally triumphant, thanks to the message of univer-
sal love and the sure promise of salvation which it offered to mankind, and
not least as a result of the strong community-organization which its first
followers set up, with unlimited charity within their ranks and stubborn
power of resistance against persecution from outside. But an epoch in which
the best of humanity sought their happiness in life beyond the bounds of
actual existence must inevitably be a period of decay, both materially and
within those spheres of the spiritual life which have to do with reality : exact
science as well as creative art.
As early as the second century of our era, when material prosperity was
still at its highest, there appear signs of this spiritual disintegration; during
this century lived the last of the great classical authors — the Latin poet
Juvenal, and the Greek Lucian, by the side of a mass of representatives of the
new era: miracle-workers, soothsayers, and necromancers, whom they
strenuously but vainly opposed. At that time, too, lived the last great biolo-
gist of the age of classical culture, the physician Galen, who in his writings
^ The cult of sacrifice was also revived, it is true, in late antiquity, but it was not so much
the old national cult as one accompanied by mystical, impassioned ceremonies, originating from
the East. To the noblest minds of the time, however, it had very little, or at any rate a purely
conventional, value.
6o THE HISTORY OF BIOLOGY
has strangely combined the many-sided biological learning of antiquity with
the mystical trend of thought of the new era.
The last great biologist of antiquity
Galen was born in 131 at Pergamum in Asia Minor, of Greek parents.
After moving to Rome he latinized his name and called himself Claudius
Galenus, but continued to write in Greek, this being a characteristic ex-
ample of the mixed culture prevailing at this time. His father, Nicon, was
an architect; through a dream he learnt that his son was destined to be a
physician, and Galen thus entered upon his medical career under what was
thought to be divine instigation. Even before this occurred, he had been
initiated by good teachers into the philosophy of his time: in his native town
he studied under Platonists, Epicureans, and Stoics, but he was particularly
versed in the writings of Aristotle and Theophrastus. Medicine he studied in
his native country, then in Corinth, and finally in Alexandria, everywhere
acquiring, besides medicine, philosophical knowledge from the best teachers
of his time. Having thus completed his education, he returned home in the
year 158 and was employed in his native city as physician in the temple of
.^sculapius, as well as, characteristically enough, at the city's gladiatorial
school. After six years, however, he moved to Rome and there began to give
lectures on his own scientific subjects, as a result of which he won the friend-
ship of men of repute and the envy of other physicians. In a still greater
degree did the immense practice he acquired awaken feelings of bitterness,
and as he himself never attempted to conciliate his envious fellow physicians,
but on the contrary strongly resented the decline in the efficiency of medical
practitioners, such a storm of hostility broke over his head that for a time
he had to fly the field and return to his own country. Wa had, however, such
an established reputation that after the lapse of onh' a few years he was
summoned to become body-physician to the Emperor Marcus Aurelius, and
thus, under the protection both of him and of his son Commodus, he was
able to carry on his work in Rome unmolested. His last vears passed quietly;
it is not known when and where he died, but probably about the year 2.10
at Pergamum.
Galen s zvri tings
As a writer Galen was very productive; he himself states that he wrote 156
treatises, of which 131 were of a medical character. Of the Ktter, 83 are still
extant. His other works embraced philosophy, mathematic?, grammar, and
law, but most of them are now lost. He was thus a many-sided man of cul-
ture, with interests far above the specialized fields of activifv of contempo-
rary physicians and even of the classical Alexandrine doctors, and well
capable of critically examining the various medical schools, which by work-
ing in opposition to one another with their dogmatically formulated pro-
grams brought medical science into disrepute. He also laid great store by
CLASSICAL ANTIQUITY, MIDDLE AGES 6l
universality of knowledge, and in one of his writings which has been pre-
served to us he exhorts his professional brethren to devote themselves to the
study of philosophy as an essential foundation for acquiring a proper con-
ception of man's nature in sickness and in health. For this purpose he refers
above all to Hippocrates, whom he extols with extravagant words in all his
works, declaring that his dicta should be interpreted as if they were the
utterances of a god. But both Plato and Aristotle also represent sources
whence he gained a true idea of nature and life, and on their ideal conception
of existence he has based his theory of biological phenomena. He has adopted
their fundamental principle of a divine intelligence as the origin and ruler
of all things, whose existence is proved by the finality of nature, and also
the theory of the soul as a purpose justifying the existence of the body. But
while Aristotle showed the finality of nature by comparing different forms
of life and pointing out the consistency displayed in their existence and
evolution, Galen deals only with the organs of the human body and seeks to
prove how in the smallest details they are constructed and applied exactly as
they should be. And in this perfect organization in the human body he sees
proofs of the power and wisdom of the Creator, whom he never wearies of
praising in words testifying to a deep personal sense of religion, and in a
tone which differs widely from the temperate scientific feeling with which
Aristotle shows the necessity for a supreme intelligence in existence. And
alternating with these pious expressions there are in his writings uncontrol-
lably violent outbursts against the representatives of the theory of the domi-
nance of necessity in nature and of the atoms as the primary components of
matter, particularly against Epicurus and his disciple the physician Ascle-
piades.^ In one other respect also Galen proves himself to belong to a new
era; not only are the old Greek philosophers quoted by him, but he also re-
fers to the Mosaic story of creation, with which, it is true, he is at variance —
in fact, he believes matter to be eternal and denies the possibility of creation
out of nothing — but which nevertheless certainly influenced his conception
of nature. That, indeed, constitutes, one might say, one single hymn of
praise to the wisdom of the Creator. In every detail of the human system
does the divine Providence show its foresight; in the hand not only the num-
ber and length of the fingers, but even every tendon and muscle is a proof
thereof; likewise with the minutest details of the rest of the body. He scorn-
fully rejects the assertion of the Epicureans that the organs develop with use
and weaken with disuse, saying that in that case energetic people would in
time acquire four legs and four arms, and the lazy only one of each. Again, if
rightly viewed, even organs which might appear to be useless are suited to
2 Asclepiades was a famous Greek physician who lived in Rome during the first century
before the Christian era. His writings, now lost, were highly esteemed in antiquity, but they do
not seem to have included any biological investigations.
6i THE HISTORY OF BIOLOGY
serve a certain purpose; if it is asked, for instance, why man has not long
ears like the donkey, which would give better hearing, the answer is that
man's ears are such as they are in order to enable him to wear a hat. And the
wise Creator has not only taken utility into consideration when making man;
He has even taken thought for human beauty, as may be seen in the distribu-
tion of hair on the face; the beard on the chin is a suitable adornment for a
man, while the growth of a beard on the nose would give to the countenance
a wild and barbaric appearance. But if the hair of the head is thus the work
of the Creator, the hairs on the arms and legs are the work of chance; they
are likened to self-sown weeds — the Greeks who went about with bare
arms and legs considered these hairs disfiguring and therefore removed
them. To such absurdities did Aristotle's theory of the finality of nature
gradually lead, its application no longer being guided by his own sober and
clear logic. However, the piety of Galen expresses itself in nobler and deeper
thoughts when he leaves anatomical details and proceeds to ethical problems.
There is a truly biblical tone about words such as these: "In my opinion
true piety consists not in sacrificing hundreds of beasts or offering quantities
of spices and incense, but in oneself knowing and learning about the wisdom,
power, and love of the Creator." Equally noble are the words with which
he exhorts his colleagues not to strive for profit, but to offer themselves to
the service of suffering humanity. In this Galen shows the same noble and
humane spirit as his lord and master, Marcus Aurelius, expressed in his Medi-
tations, and there is every indication that, like the latter, he lived as he
taught.
Galen's anatomical investigations
But if Galen in his general conception of life thus stood on the border-line
between antiquity and the Middle Ages, he was as regards knowledge of
anatomical detail the foremost philosopher of the classical period, and as
such remained the undisputed authority in his own branch of learning up to
the Renaissance, and strictly speaking even up to the time when Harvey
discovered the circulation of the blood and thereby destroyed one of the
foundation-stones of his theoretical system. Both as an anatomist and as a
physician Galen had indeed the inestimable advantage of being able to build
upon the work of brilliant predecessors, but he also realized the importance
of his inheritance and considerably enhanced it by his own observations.
These he carried out exclusively on animals, both dead and alive, especially
apes, which he considered particularly suitable as material for investigation
of human anatomy. There is never any question of his dissecting human
bodies; the times had changed considerably since the days of Herophilus;
old superstitions had again been revived and governed men's minds. He
characteristically begins his enunciation of the anatomy of the human body
with the hand, the most useful of all organs — that whereby the soul effects
CLASSICAL ANTIQUITY, MIDDLE AGES 63
its will, for the whole body exists for the sake of the soul. The human hand
is described in detail and with great thoroughness, but, as one can clearly
see, as the result of investigating the hands of apes. Then he describes the
rest of the extremities and afterwards the intestinal canal, the respiratory
organs, the brain, the spine, the blood-vessels, and the genital organs. Galen
stands highest as a brain and nerve anatomist, and in this sphere his anatomi-
cal and experimental investigations gave results which left all his predeces-
sors far behind. He considerably increased the knowledge of the motor and
sensory function of the nerves, which the Alexandrine anatomists had already
observed, and he differentiated between the sensory, or, in his terminology,
the "soft" nerves, and the motor, or "hard." The soft nerves go from the
brain to the sense organs, the hard from the spinal marrow; as the nerves of
the spinal marrow also show definitely sensible qualities, though Galen did
not succeed in discovering the difference between the anterior and the pos-
terior medullary nerves, he evades the difficulty by assuming a "mixed"
consistency and function in certain medullary nerves. By experiments in sever-
ing different sections of spinal marrow in living animals he showed the con-
nexion between these and corresponding parts of the body. The brain he
likewise described in detail; of its nerves he traces seven couples, the rami-
fications of which are closely worked out. On the other hand, his idea of
the function of the brain is confused, owing to speculations upon the "soul
pneuma," which, produced in the cerebral ventricles, circulates through the
entire nervous system and forms its most essential component, the basis of its
functions. It may be mentioned in this connexion that he shared the ancient
idea of the localization of the various qualities of the soul in various organs,
which naturally gives rise to long expositions on the wisdom of the Creator
and the finality of the creation. The account of the digestive apparatus and
its function is in Galen, as in the anatomists of antiquity in general, one of
the weak points. The human digestive canal is described after combining
the results of dissections performed on various animals, both vegetarian and
carnivorous, which fact does not help to make his idea of it clear. Digestion,
which in Aristotle was the result of the cooking of the food, is ascribed by
Galen to a special "transformation power" in the stomach; its products are
transferred through the blood-vessels to the liver, where they are converted
into blood; the useless parts of the food are absorbed by the spleen and con-
verted by it into "black bile," which is excreted through the bowel. The
kidneys serve to remove excessive water from the blood. This is afterwards
conveyed through the veins of the liver, partly to the right chamber of the
heart and partly out into the body.
Heart and blood-vessels
Galen described the system of the blood-vessels in detail, and his opinion
on this subject — with its errors as well as its merits — had a more lasting
64 THE HISTORY OF BIOLOGY
effect than that on any other subject. As one of his services to science it may
be mentioned that he finally succeeded in overcoming the old preconception
that the arteries and the left heart-chamber contain air. In his opinion they
contain blood, with an admixture of "pneuma" — that half-airlike, half-
firelike life-principle which gave rise to so much discussion in ancient times
and on which the existence of living creatures depends. In one place Galen
expresses the hope that the time may come when someone will discover the
component in the air which forms pneuma, the substance which is the com-
mon precondition of life and combustion — a curious idea, which in fact
the discovery of oxygen was eventually to bring to realization. He gives a
detailed description of the heart, both its structure and its functions; on the
other hand, like his predecessors, he lets both veins and arteries convey the
blood from the heart to the rest of the body, in which it is consumed. He
is not aware of any blood flowing from the body to the heart, while his idea
of the movement of the blood is still further confused by his belief that the
liver is to a certain extent the centre of the venous system, since the blood
flows from it not only to the heart, but also to the rest of the body. The
left ventricle receives through the vena pulmonalis "pneuma" from the lungs;
from the right ventricle the excremental products proceed to the lungs, these
products being "soot" from the combustion process in the heart, which is
got rid of by exhalation. The wall between the right and the left ventricles
of the heart is porous, permitting the blood to pass through. The walls of
the blood-vessels are carefully described, and, generally speaking, Galen's
detailed study of the construction of the individual organs is one of his strong
points. He is aware of the connexions between the arteries and the veins,
but, as is seen from the above, he has not realized the idea of circulation, and
this fact, combined with the vagueness with which he explains his ideas on
these organs, proved an obstacle to the development of biology for the next
fifteen hundred years.
Galen carefully studied the respiratory process and on the whole de-
scribed it correctly. With regard to the sense organs, in spite of his thorough
investigations into the subject he made very little advance on his predeces-
sors, and the same may be said of his description of the genital system and
the embryonic process, in which he remains, on the whole, where Aristotle
stood.
Splendid as Galen's scientific work was, it does not appear to have been
highly appreciated by his contemporaries. He himself complains that but few
understand him, but consoles himself with the thought that the Creator, in
spite of man's ingratitude, never wearies of doing good. Here posterity has to
an unusually generous extent made up for what his contemporaries failed to
give him. The fact that Galen did not impress his contemporaries may have
been due to the peculiar transitional attitude he adopted; to the survivors of
CLASSICAL ANTIQUITY, MIDDLE AGES 65
the purely ancient school his romantic piety must have been repulsive, while
the more mystically minded, who even then represented the majority, were
on the whole not at all interested in exact natural science. The miracle-
workers' laying on of hands and invocations were undoubtedly more relied
upon than Galen's curative method, based on anatomical studies. On the
whole, during this period interest in the study of nature waned more and
more, at least in the sense in which the philosophers of old times understood
it; the most one could do for educational purposes was to collect stories
about natural objects. One such collection is the treatise still preserved in
our day On the Habits of Animals, which was written by Claudius v^lianus
a generation after Galen. This writer, who was an orator by profession —
that is to say, a public lecturer — lived in Rome in the first half of the third
century; he is believed to have died in the year x6o. His work is a collection
of anecdotes about animals, gathered from various sources — thus following
the method of Pliny. But while in Pliny the interest in nature is the prin-
cipal motive, i^lianus is actuated by a feeling of pure edification. Pliny, it
is true, can also edify his readers with the examples he cites of the virtues
of elephants, but they are related in order to testify to the animals' great
qualities of soul. In .^lianus even the lowest creatures are uplifted by a
purely personal reverence for the Creator, so that the ecclesiastical writers
of the Middle Ages had only to substitute the names of Christian saints
for those of the gods quoted and they thus found ready to hand a collection
of the most edifying sermons. Thus i^lianus tells of a cock that had one
of its legs broken; the bird hopped on its other leg before a statue of a god,
and stretching out the broken foot, crowed so pathetically that the god
showed his mercy by miraculously healing the injury, whereupon the cock,
gratefully flapping its wings, went on its way. Here we find the purely
mediaeval conception, and this more than a century before the final victory
of Christianity; an example, i^iter alia, of the incorrectness of the frequent
assertion that the Christian Church after its victory eradicated the culture
of antiquity.
Neo-Platonists
With regard to the natural sciences an attempt has been made in the fore-
going to throw light on the process of internal dissolution which gradually
led biology from Aristotle's magnificent system of thought down to i^li-
anus' collection of legends. The interest in natural phenomena which had
for so long been a living factor in the ancient world of culture had now
entirely disappeared. What was left of the spirit of research turned to ideal-
istic philosophy, Plato's creation, which was further developed by thinkers
who adopted his name and made his theory of ideas their starting-point,
proceeding thence in a curious direction, at the same time speculative and
full of religious mysticism. In their relations with the outer world these
66 THE HISTORY OF BIOLOGY
neo-Platonists were as opposed to the men of antiquity as their Christian
contemporaries. The founder of the school, Plotinus, was ashamed of pos-
sessing a body, while its last great thinker, Proclus (411-485), lived as a
hermit; he dwelt in a cave, avoided wine, meat, and women, and saw visions
of supernatural things. For to be uplifted into the supersensible by means
of ecstatic rapture was to the neo-Platonists not only the whole object of
life, but also the very foundation of science. With all its fantastic specula-
tion this school nevertheless developed human thought in one important
sphere; it discussed the idea of infinity as none of its predecessors had been
able to do. To the ancient atomists infinity was really only an unlimited
extension in time and space, akin to the custom of children and wild men,
who when they are weary of counting, call the remainder "much" or
"many." To the neo-Platonists, again, the infinite was equivalent to the
inexpressible and the unknowable, that which exceeds all limitations and
measures. And though their endeavour to attain to this infinity by way of
ecstasy was naturally of no scientific value, there was nevertheless an in-
disputable truth to be gained as a result of their endeavours, seeing that
the impotence of the power of knowledge in face of the infinite was estab-
lished once and for all. The natural-research work of our time is based on
the realization of the limitation of knowledge in face of the infinity of
existence — a limitation which only unscientific dilettantism thinks it pos-
sible to override.
Destruction of the old culturt
There were, then, among the thinkers of this period ideas which pointed
beyond the limitations by which the ancient conceptions of existence had
been surrounded. It is impossible to estimate how these aims might have
developed in happier external circumstances. For as a result of the fall of
the Roman Empire the external, purely material preconditions for the con-
tinuance of scientific research and for the progress of culture in general no
longer existed. As early as the latter half of the imperial epoch the pros-
perity of those nations which formed the Roman Empire steadily declined
as a result of misgovernment, civil war, and the inroads of neighbouring
peoples. In the fifth century the world empire collapsed entirely owing to
the invasion of the Germans, and a state of dire distress, economic, politi-
cal, and moral, ensued. The new kingdoms which were founded by barbar-
ous nations had great difficulty in firmly establishing themselves, and their
rulers' utter lack of culture rendered impossible any kind of ordered system
of government and, consequently, any high standard of prosperity. How-
ever, the inhabitants of western Europe gradually co-operated in reviving
culture on a national basis. During the last hundred years of the Roman
Empire, Gaul had been the most civilized country in the Empire, with
numerous institutions founded for the study of classical learning. During
CLASSICAL ANTIQUITY, MIDDLE AGES 67
the period of invasion many people fled from western Gaul to what was
certainly a barbarous, but all the same a peaceful country, Ireland, and thus
was founded a centre of culture which during the sixth and seventh centu-
ries was the foremost upholder of the classical tradition and one of the
starting-points for the future progress of civilization. In the eastern part of
old Roman Empire the Byzantine power was still dominant with a despotic
form of government and the Greek Orthodox Church as a binding force.
There, too, efforts were made to develop national culture, which was ex-
pressed in literature, in the national tongue, combined with interest in
Greek science. This was especially so in Syria, where the national move-
ment was often associated with religious sectarianism; but in Persia too,
under the Sassanid dynasty, Greek science was studied, particularly Aris-
totle. In these countries, however, there shortly arose a new cultural
power, which took over and further developed the learning of the classical
period — namely, the Arabian people, with their new religion, founded by
Mohammed.
CHAPTER IX
BIOLOGICAL SCIENCE AMONG THE ARABIANS
The Arabian conquest of the East
WHEN MOHAMMED DIED, iH 63X, the religion he founded had already
spread throughout Arabia, and his successors, the first caliphs,
managed in the course of a few decades to bring under their do-
minion the old civilized countries of Babylon, Persia, Syria, and Egypt, to
which were later added North Africa and Spain. War against the unfaithful
was indeed the prophet's first commandment, and according to his injunction
the heathen had a choice between death and conversion; such of the unfaith-
ful, again, as possessed religious writings — Christians, Jews, and Persians —
had their lives spared, but were subject to impositions and personal humilia-
tion. The bedouins of the desert, who thus at one blow became the rulers of
the most ancient civilized countries in the world, were themselves nothing
but barbarians, it is true, but they were intelligent and susceptible to cultural
influence, all the more so as in the course of the wandering life they led, they
had already come into contact with their civilized neighbours. Their new re-
ligion was favourable for rapid cultural progress in that it was a legal doc-
trine with few and easily comprehensible rules, without, to be sure, the lofty
ethical claims of Christianity, but also without the theological subtleties of
the different ecclesiastical formulas. And as, besides, the Arabs troubled them-
selves but little about social and political questions — they permitted the
institutions of conquered nations to survive and contented themselves with
appointing governors who collected taxes from them — they had ample time
to devote themselves to purely intellectual interests. Indeed, they grasped the
elements of the culture of the period with a rapidity which has been com-
pared to that of the Japanese in our own day, and were able in many respects
to build higher upon the foundations they found prepared for them. These
foundations were Greek science, as the subject peoples produced it in Syrian
and Persian translations; it was not until later that the Arabs learnt to read
Greek writings in the original. They developed this material and thus created
a science representing at the same time a direct continuation of the Greek
and a reconstruction of it to suit the conditions which the peculiar Arabian
view of the world required. According to Mohammed's theory, the Koran
is the source of all learning and contains all the knowledge that man requires;
but this claim, which would have rendered all research impossible, was
68
CLASSICAL ANTIQUITY, MIDDLE AGES 69
evaded during more liberal eras, while obscurantist rulers continued to
threaten the learned with its literal application. This was, however, the
cause of a certain restraint invariably characterizing Arabian research, at least
in form; the scientists preferred to give their works — even the most independ-
ent — the appearance of commentaries on the writings of some famous scien-
tist of antiquity. In philosophy and natural science it was naturally Aristotle,
in medicine Galen, who was made to represent the authority on whom
the work was based, and at the same time the screen behind which the Ara-
bian scientists saved themselves in the event of the authorities' finding the
results of their research work inadmissible. During the most brilliant period
of Arabian research it was certainly possible for original and great thoughts
to be disguised beneath these commentaries on ancient writings, but the
danger of slavish imitation lay in the method itself, and for more than five
hundred years the science of the East was drowned in an utterly soulless
amplification of ancient authorities.
Experimental method introduced iyito science
The Arabian contribution to the development of the exact sciences has been
most important in the spheres of mathematics and astronomy, in which they
received impulses not only from Greek, but also from Hindu quarters — the
so-called "Arabic numerals," which are now universally used, were bor-
rowed by the Arabs from India — and, further, geography, a study which
the Arabs applied to the investigating of several unknown regions, medicine,
particularly pharmacology and, in connexion therewith, botany, and, finally,
chemistry, which they were the first to raise to the rank of a science. Chem-
istry, indeed, is experimental science above all others, and with it experi-
menting as a scientific method was introduced and developed by the Arabs.
This contribution alone is such as to ensure to Arabic science a place of
honour in the history of research. Experimenting, in which the research-
worker himself interferes with the course of events in nature and arranges
that course with a view to having a specific question answered — this, the
most certain method whereby the obedience of natural phenomena to law
can be proved, was unknown to ancient research. Even Archimedes himself
was no experimental physicist, eminent though he was as a practical en-
gineer, while Galen's vivisections, as well as those of his Alexandrian prede-
cessors, had rather the character of observations of live animals than of
actual experiments. As a matter of fact, however, the experimental method
is very ancient and has its origin in a number of experiences of various kinds
which survived in different classes of people before science adopted their
methodic system and employed it for obtaining results in exact research
work. Thus every type of peoples has practised magical experiments based
on the preparation of charms, which are concocted out of the most extraor-
dinary ingredients and are used as love-potions, elixirs of life, enchant-
yo THE HISTORY OF BIOLOGY
ments, and pure poison. Such magic brews were prepared among the nations
of antiquity by witches and wizards and are still concocted amongst inferior
peoples and even amongst more primitive strata of higher types to this very
day. That the enlightened scientists of antiquity refused to associate them-
selves with such magical preparations was but natural; it required that
tendency towards the vulgar and the fantastic which the decline of ancient
culture evoked in the scientific world, before the methods of popular sorcery
could begin to be of interest to thinking and inquiring minds as well. Nor
indeed does the earliest experimental science deny this origin; it appears in
the form of alchemy, with its pronounced mystical aims, and above all in
the conversion of base metals into precious metals, the discovery of elixirs
of life and immortality, the reproduction of homuncules, etc. — aims to
which it adhered throughout the Middle Ages, even after its means and
methods had become characterized, at least in certain features, by a fair
measure of exactness and an extensive knowledge of the inorganic objects of
nature in particular. It was therefore at a later stage than any other branch
of exact science that the experimental branch succeeded in freeing itself from
its connexion with the supernatural world of thought, from which all science
gradually broke away. It is only the research work of more modern times
that has been able to enjoy to the full the advantages which experimental
science offers.
Arabian natural philosophers
With biology the famous Arabian alchemists, one Geber and others, had
nothing to do; they occupied themselves only with inorganic nature. On
the other hand, the East possessed a number of purely speculative re-
searchers who dealt with the phenomena in living nature from a theo-
retical point of view and who exercised a lasting influence on the conception
of them in succeeding ages. All these philosophers took Aristotle as the
starting-point for their researches and, as already mentioned, they likewise
gave to their own often quite daring speculations the form of commentaries
on his works. Indeed, their position was always fraught with danger; they
were looked upon with suspicion by the orthodox Mohammedans, who be-
lieved that all studies that did not concern the Holy Koran were prohibited.
Against these constant persecutions they had no other support than the pat-
ronage of some science-loving prince, which had to be won and sustained by
flattery and was at best an unreliable guarantee of life and maintenance.
These philosophers held no posts as teachers — in the Mohammedan East
there were colleges only for students of the Koran — but their scientific
researches were always a private occupation; by profession they were fre-
quently physicians, sometimes lawyers, officials, or courtiers.
Among these oriental thinkers there are primarily two who exerted
some influence on the progress of science even in the West, their writings
CLASSICAL ANTIQUITY, MIDDLE AGES 71
being at an early date translated into Latin and diligently studied in Europe,
becoming at the same time the basis for continued philosophical research.
Abu Sina, or, as he is called in Europe, with a latinized distortion of his name,
AviCENNA, was born at Bokhara in 980, of Persian stock. At that period
Persia was divided into a number of major and minor states, ruled over by
princes, who in mutual rivalry sought to win honour by exploits of war and
peace. There prevailed a high standard of intellectual culture, and the con-
ditions of the country have often been compared with those of Italy during
the Renaissance. Avicenna indeed bears a great resemblance to personalities
living at the time of the Renaissance; strictly speaking, he was a physician,
but he was also mathematician, astronomer, philosopher, and poet. Cheq-
uered too were his fortunes; at one time he was an all-powerful minister at
the court of some vassal prince, at another he was an exile fleeing from his
enemies and in danger of his life. He died in 1037, his health shattered by
his manifold exertions and his reckless love of pleasure. The most important
of his numerous writings is his great Canon of Medicine, which, next to
Galen's, remained the chief authority in the sphere of medical science. Its
sections dealing generally with natural philosophy, anatomy, and physi-
ology are of interest from the point of view of biological history. There is
still extant a major work of his on general philosophy. As a thinker Avi-
cenna takes Aristotle as his starting-point, but he is also tq. a certain extent
influenced by neo-Platonism. His conception of nature is governed by the
"purpose" theory of Aristotle and Galen. Entirely based on Galen, too, is
his idea of the human anatomy. The Arabs were in fact even more afraid of
dissecting human bodies than were the people of antiquity; it was forbidden
in the Koran, and with however little prejudice the learned interpreted the
sacred book, they dared not in this respect violate both it and public opinion.
Avicenna, however, was more independent as a physiologist; here he could
take advantage of the progress his contemporaries had made in the fields of
physics and chemistry. But actually it was more for his excellence of form —
brilliant style and well-arranged grouping of his subject — than for any
original ideas that Avicenna won fame in the East and eventually, perhaps
to a still higher degree, in the West.
Far more original as a thinker is the second of the great men of science in
the East, Averroes, or Ibn-Rushd, as he was properly called in Arabic. He
was born at Cordova in Spain in iiz6, the son of an eminent judge. In his
native city, which for several centuries had been the centre of Arabic culture
in Spain, he studied philosophy, medicine, and jurisprudence, was for some
years afterwards cadi of Seville, and was finally governor of a province. The
fanatical religious reaction, however, which gradually spread among the
Mohammedans in Spain towards the close of the twelfth century, once suc-
ceeded in bringing about his downfall, and, accused of heretical opinions
72. THE HISTORY OF BIOLOGY
and pursuits, he was imprisoned, stripped of his honours, and banished to
a village near Cordova inhabited by Jews. Fortunately the ruling prince who
committed this act of injustice died a year or two afterwards and his son
and successor immediately repaired it; Averroes was recalled to court and
resumed his honours, but died shortly after, in 1198.
As a natural philosopher Averroes followed Aristotle, and his principal
work takes the form of commentaries on Aristotle's writings. Averroes's
standpoint is, however, far more than that of his predecessors and even than
that of any other mediaeval philosopher, independent of his model. He bases
his philosophy on the latter' s ideas, but he develops them further on his
own account. In particular he studied the relation between potentiality and
reality in nature. Aristotle considered that marble is a potentiality, which
becomes reality when a statue is made out of it, and he consistently applied
this method to life in nature — the seed, the embryo, is a potentiality; the
plant, the animal, reality. Averroes argued in opposition to this view that
nothing in nature is potential that does not exist in reality, in however un-
developed and therefore disguised a form it may be; the plant already exists
in the seed, in however undeveloped a state, just like the animal in the embryo.
The simile of the marble and the statue Averroes considers inapplicable where
nature is concerned; at best the simile would be admissible if the statue were
to be found already shaped in the veins of the unsculptured block. By this
method of speculation Averroes has carried science a long step nearer the
present-day conception of natural evolution; Aristotle's purely abstract
idea of potentiality is here replaced by something which approaches far
nearer to our idea of energy. Averroes was the last great Arabic philosopher
and the greatest natural philosopher of the Middle Ages; if anyone is worthy
to be called the Aristotle of the Middle Ages, it is he. He resembles his pro-
totype not only in the fact of his having lived in a decadent era and been
subjected to religious persecution, but also in the fact that no one for cen-
turies succeeded in developing his ideas further. Shortly after his death
Arabic science succumbed to religious intolerance, while even the Christian
schoolmen, who closely studied and highly honoured the Arabic philoso-
pher,^ saw in him only the interpreter of Aristotle and were not capable of
realizing the great advance he made towards a more real conception of na-
ture. He has not, however, been without influence; in the Middle Ages the
opponents of ecclesiastical philosophy gathered round his name, and the
ideas he evoked can thus be traced through the ages until they find confirma-
tion in the natural science of our own day.
Arabic literature has produced, besides the natural philosophers men-
tioned, several authors who have dealt with zoology in a more restricted
^ During his imaginary wanderings in the underworld Dance sees Averroes in the court
of the heathen by the side of Aristotle and other philosophers of antiquity.
CLASSICAL ANTIQUITY, MIDDLE AGES 73
sense: faunistics and zoogeography. Such authors are mentioned as early as
the ninth and tenth centuries, but their writings have not been preserved.
On the other hand, there is still extant an account of the animals of Egypt,
written by Abdallatif (1162.-1131), which is manifestly based not only upon
ancient authors, but also upon personal investigation. Inter alia, he gives
a detailed description of the hippopotamus and the crocodile and also an
account of the method customary in Egypt of hatching hen's eggs by arti-
ficial heat. A fairly large work entitled Animal Life by Muhammed el Damiri,
written at the end of the fourteenth century, has come down to us. He has
described a great number of animal species — one statement declares it to
be nearly nine hundred — partly based on his own observations, but partly
also on pure imagination. Arabic literature possesses one author comparable
with Pliny in the person of Sakarja ben Muhammed, called el Kasvini after
his own district of Kasvin in northern Persia. He lived in the thirteenth
century and thus had at his disposal, besides Aristotle and Hippocrates,
whom he freely quotes, a number of Arabian predecessors, of whose works he
made extensive use. His important collective work. The Wonders of Nature, is
based on Aristotle's natural philosophy of evolution from the lower to the
higher; the capacity to feel and move differentiates the plants from the ani-
mals. His theory of fossilized animals is curious; he believes that they have
been petrified by steam arising out of the ground on which they stood. For
the rest, he describes a number of tropical animals which were unknown to
ancient authors, for instance the orang-utan, which he pictures as having
the human characteristics which the inhabitants of his native place ascribe
to it, and, further, the flying dog, the dugong, and others.
On the whole, it is through their having promoted the knowledge of
and the cultural influences between the East, even its most distant parts, and
the West, that the Arabs have become best known among the peoples of the
West; rather than by the really more profound cultural service they performed
in having preserved and developed the remains of ancient culture at a period
when the West was incapacitated from preserving the inheritance which
nevertheless most directly devolved upon its peoples. Through the interme-
diary of the Arabian philosophers the few learned scholars of the West in the
early Middle Ages acquired a knowledge of the products of classical culture;
Aristotle, for instance, was long read at the mediasval universities in Latin
versions of Arabic translations from the original writings, and the Arabic
commentators, Avicenna, Averroes, and others, were the first to act as guides
to an understanding of the treatises on nature and to help Europeans to pene-
trate that world of phenomena whose existence they had entirely forgotten.
Thanks to Arabian science, the so-called dark centuries of the Middle Ages
were at any rate culturally fruitful, and when oriental science, after flourish-
ing for a brief period, died out, the people of the West had already laid the
foundations of an entirely new cultural development.
CHAPTER X
BIOLOGY DURING THE CHRISTIAN MIDDLE AGES
The Eastern Roman Empire
IN A PREVIOUS CHAPTER mention has been made of how the culture of an-
tiquity, itself already become decadent, received its death-blow through
the transmigration of peoples which broke up the Roman Empire. The
first political evidence of this dissolution was the splitting up of the mighty
Empire in the year 395. The cultural world of the time was thereby divided
into an eastern and a western half, which suffered essentially different for-
tunes. In the eastern section the old imperial constitution was still to survive
for over a thousand years, maintained in power through the people's being so
long accustomed to a despotic form of government, and upheld by an intimate
connexion with the strangely established Greek Oriental Church — in actual
fact the bond that held together the mixed populations which gave alle-
giance to the sceptre of the Emperor of the East. Greek was the prevailing
language here and the medium for a peculiar form of culture, the Byzantine,
which displayed extraordinary qualities of resistance to the pressure of hos-
tile forces: the Mohammedans in the East, wild hordes of migratory peoples
from the north and "Latins," as the western Europeans were here called,
in the West. This constant struggle for cultural supremacy produced, as it
invariably does, a tendency to strict conservatism, and the value of the By-
zantine culture therefore lies not so much in independent creative work as
in all that it did for the preservation of the ancient literature which, even
for philological reasons, it already had some interest in preserving. The capi-
tal of the Empire certainly possessed valuable libraries, and educational es-
tablishments with highly complicated methods of instruction, but the studies
pursued there consisted mostly in theological subtleties, the amplification
of ancient authors, and the compilation of histories. The scholars of Con-
stantinople cared little for natural science. On the other hand, the Byzan-
tine physicians were famed for their great ability; they honourably upheld
the best traditions of the medical science of antiquity. Their training was
entirely practical, however — they received no academical instruction in the
science of medicine — and they were in fact essentially practitioners; the
theoretical branches of medicine, anatomy and physiology, they have done
very little to promote. The principal medical work of the Byzantine era,
written by Paulus of^^gina in the seventh century, deals only with practical
74
Q
<
>
o
p
<
w
1/5
I— I
h-l
o
w
w
<!
>
o
H
CLASSICAL ANTIQUITY, MIDDLE AGES 75
medicine; its surgical section is celebrated for its excellence and has had
great influence on the medical science of both Arabia and the Occident. —
The Byzantine Empire and its culture eventually succumbed to the Turks,
but before that it had had time to exercise considerable influence upon west-
ern European civilization, especially by spreading a wider knowledge of
classical Greek literature and thereby paving the way for the great cultural
regeneration of the Renaissance.
Western culture
The western Roman Empire, unlike its eastern neighbour, fell a prey to
hordes of migratory peoples and was dissolved by them into a number of
minor states with constantly changing frontiers and unsettled internal con-
ditions. The only one of the kingdoms founded under these circumstances
which attained a successful development was that of the Franks, which at
one time, under Charlemagne, embraced a large part of the western Roman
territory and more as well. After his death his empire fell to pieces and out
of its ruins gradually arose the national states of western Europe which still
exist today. During the centuries of migration both material prosperity and
intellectual culture in the western Roman countries were destroyed. The last
remains of classical culture found a refuge in Ireland; there in the sixth and
seventh centuries were read and copied not only Latin, but Greek authors,
and thence culture spread to England, at that time conquered by the Anglo-
Saxons. In Charlemagne's time these two countries possessed the highest
intellectual development and it was from there that the Emperor summoned
learned men, with whose aid he raised the standard of culture in his own
country and created what was called a "renaissance" in the field of classical
studies. After his death, however, western Europe was ravaged by a fresh
barbaric invasion by the Danes, which destroyed culture exactly where it
had hitherto been most highly developed -^ in Ireland, England, and France.
The most decadent period of the Middle Ages really set in during the ninth
and tenth centuries, just when the Arabic culture was most flourishing.
The one power that kept men together in that unhappy period was the
Catholic Church; it gave consolation and support in time of trial and was
able to induce minds broken down by misfortune to strive after ideals. As a
unifying cultural force it came to take the place of what the Empire had once
been, and so Rome became once more the capital of the world. But while
the Church thus gave to culture fresh vitality, it at the same time set rp nar-
row limitations for its development; it demanded absolute subjection, to
the extent that not only did religious sentiment have to choose the paths
the Church prescribed, but even the human intelligence had to adhere to its
dogmas and doctrines as proved truths. These had been drawn up by the ec-
clesiastical Fathers of the first centuries of the Christian era, whose writings
the priests and monks of the early Middle Ages read without interpreting
76 THE HISTORY OF BIOLOGY
them or adding anything fresh to them. Not until the latter half of the
eleventh century did the first independent mediasval theologian appear, in
person of Anselm of Canterbury. But not long afterwards a more liberal
line of thought began to find expression, which, based on the trifling remains
of classical literature still to be found at that time in the libraries of monas-
teries and churches, sought to establish rational principles of thought. Dur-
ing the twelfth century these ideas, expounded by the Frenchman Abelard
and his pupils, won widespread acceptance, in spite of strenuous opposition
on the part of the Church, and received further stimulus from the influence
of Arabian science, brought over partly by scholars who had studied in Spain
and partly through the crusaders' contact with the East itself. In this way
the countries of the West gained their knowledge of the great men of classical
antiquity — - Plato and Aristotle, Hippocrates and Galen, as well as of their
Arabian commentators and successors.
The universities
With a view to the study of these sources of knowledge there was founded
in the twelfth century a form of educational establishment which was to
become of fundamental importance for the scientific development of the
future — namely, the university. Antiquity had nothing equivalent to this
kind of associations of teachers and pupils, for they rested on an ecclesias-
tical foundation. Charlemagne had already founded and attached to the met-
ropolitan churches cathedral schools, in which some young priest gave
instruction in theology, music, and other branches of learning necessary for
men of the Church. As the number of pupils at these schools gradually in-
creased, it became necessary to employ in them a larger and larger staff of
teachers, magistri, who, in order to protect themselves against the dangers
and insecurity that prevailed everywhere in those days, formed themselves
into corporations. An association of this kind, universitas magistrormn, under
its governor (j-ector) and with its large number of pupils grouped according
to nations, represented at that epoch a considerable power, which, in the
course of violent struggles with the civic and ecclesiastical authorities, en-
deavoured to acquire a wide measure of self-government and as a rule actually
succeeded in doing so. When the number of both pupils and educational
subjects increased still further, recourse was had to specialization in several
faculties, a method of distribution which in its main features still survives.
Instruction was given by means of pulpit lectures — a method based on the
Church sermon and similarly adopted with a view to instructing large num-
bers of pupils at one and the same time. Further, the appearance of the
university system involved a democratization of science, of which classical
antiquity had no counterpart. Whereas the finest masters of antiquity could
probably count their pupils only in tens or hundreds, the great universities
of the Middle Ages, such as those of Paris, Oxford, Leipzig, etc., had thou-
CLASSICAL ANTIQUITY, MIDDLE AGES 77
sands and even tens of thousands of students attending at one time. True,
both mass education and self-government had their dangers; reactionary
intellectual movements, equally with earnest strivings after knowledge,
might, and in fact did at times, gain the mastery at the medixval universi-
ties, just as, indeed, later liberal and reactionary aims alternately dominated
university research and instruction.
Scholastic doctrine
The science taught in mediaeval schools and universities — scholastics, as
it was called — was governed, as already mentioned, by ecclesiastical doc-
trine. The intellectual movements which were set on foot independently
thereof and which were consequently persecuted as heretical were not
founded, as they generally are nowadays, on natural science, but took
their stand on purely speculative ground. The question which had been de-
bated ever since the days of the Church Fathers of the relation of reason to
faith, or, in other words, the right of the individual to criticize Church
doctrine, was answered in the first place by the universities along fairly
liberal lines, in spite of protests from the Church, but in the thir-
teenth century a religious reaction set in, evoked by the struggle against
heresy and represented by the orders of mendicant friars founded for the ex-
press purpose of combating the heretical movement; finally these orders
succeeded in usurping the control of university education, at least of theo-
logical instruction, which was thus compelled to adapt itself to ecclesiastico-
political aims. This occurred, strangely enough, just at the time when a
closer knowledge of Aristotle began to be disseminated in the universities,
based on the Greek original and not merely on the Arabic translations of
his writings. But the High-Church theologians who now held sway in the
universities soon began to realize what a splendid ally they had in the Aris-
totelean philosophy, which they had originally mistrusted as mere heathen
delusion. Aristotle's conception of the earth as the centre of the universe
and yet as the home of all imperfection in contrast to the perfect heaven
might very well be adapted to the Church's doctrine of sin and salvation.
His strictly formalistic cosmic system and mode of thought, with its domi-
nating intelligence and its denial of any material causality, was, like his
conservative and authoritative view of hum^n life, well suited to form a
scientific basis for the hierarchical aims of the papal power. And if his writ-
ings did not agree in every detail with the revealed Word, the inconsistencies
made apparent thereby could be explained away by reference to the author's
paganism and ignorance of the way of salvation. Thus was created in the
thirteenth century, mainly on the initiative of the greatest thinker of the
Catholic Church, the canonized Thomas Aquinas, the curious and, in its way,
fully elaborated system of thought which that Church has ever since then,
held to be the only true one. According to this system, existence is divided
78 THE HISTORY OF BIOLOGY
into three "kingdoms," those of nature, grace, and blessedness. In the first
dwell all men; the two latter are attainable only by members of the Church.
Knowledge of nature, therefore, even the heathen may acquire, and no
heathen has possessed deeper insight in this respect than Aristotle; he has
explored the kingdom of nature with unexcelled wisdom. Consequently the
Christian researcher may safely rely upon his explanation of nature and need
not engross himself in the subject — all the less so as the kingdoms of grace
and blessedness are open to him, the former in this life, the latter in the
eternal hereafter. In these circumstances the thinkers of the Middle Ages
devoted but little attention to natural-scientific research; they contented
themselves with the writings of Aristotle, which were closely commented
upon, even down to the smallest detail, without any effort's being made to
develop their subject-matter by actual investigation. There is a well-known
story of how the learned ecclesiastics disputed as to how many teeth the
horse should have according to Aristotle, instead of looking into the mouth
of a live horse to see for themselves. So much the more to the point were the
Aristotelean problems regarding the relation of ideas to reality; here the
dispute waxed hot between the realists, who believed that ideas existed
before things, and the nominalists, who declared that ideas exist only in
things. The view of the former was eventually given official sanction, but
their opponents refused to give in and so played their part in undermining
the reputation of the High-Church philosophy towards the end of the Middle
Ages.
There are no biological writings proper dating from the earlier Middle
Ages. The descriptive work on animals, the Physiologus, which is mentioned
in all zoological histories, can indeed hardly be included in this category; it
consists of a collection of edifying stories relating to the animal world,
intended to serve as examples for quotation in sermons and gathered together
from all quarters. Probably it dates from later antiquity, which produced
many such collections, as, for instance, that made by i^lianus mentioned
above. The Physiologus, which was an anonymous treatise revised and issued
in various editions, had a surprisingly wide circulation; it was translated
into Ethiopian, Icelandic, and most languages existing between these bor-
derlands of Christian culture. It abounds in fantastic stories; a number of
them have survived even to the present day.
Even in the Middle Ages, however, there existed people who had a
broader view of nature and a deeper interest in the life that stirs therein than
had the ecclesiastical legend-writers. Interesting evidence of this is to be
found in a treatise dating from about 11 50 entitled Physica, written by the
nun HiLDEGARD,of Bingen on the Rhine. The book contains notes on animals,
plants, and stones and on the benefit that man can derive from them. It is
entirely popular in style and without any pretension to learning, and for
CLASSICAL ANTIQUITY, MIDDLE AGES 79
that very reason it is of interest as a sample of the ideas about natural objects
which people entertained in those days.
We find an author of another type in the person of the renowned Emperor
Frederick II of Hohenstaufen. As is well known, he was one of the most
remarkable rulers of the Middle Ages, Italian in his upbringing, half oriental
in his habits and mode of thinking. In his south Italian kingdom he gathered
round him learned men from the East and West. He had Aristotle's writings
translated from the Greek into Latin and founded a school of medicine at
Salerno, where for the first time since Alexandrine days human bodies were
dissected. He himself wrote a book, still extant, on falconry, a sport to which
princes and nobles were passionately addicted. Frederick's treatise is far more
than a mere dissertation on hunting; in a lengthy introduction he gives an
account of the anatomy of birds, in which he not only displays a knowledge
of Aristotle's anatomical writings, but is also able to point out inaccuracies
in his statements; further, he describes the habits of various birds, the move-
ments of migratory birds, etc. Unfortunately Frederick lived during the
period of ecclesiastical reaction in the thirteenth century, and after his death
his Church opponents eradicated most of the cultural progress he had
achieved; the dissection of human bodies was again prohibited and physi-
cians had henceforth, as before, to rely on the classical authorities. The
translation of Aristotle which he caused the learned Michael Scotus to
carry out was perhaps the most enduring evidence of his cultural aims; it
was on this work, in fact, that the scientists of the later Middle Ages in
general based their learned studies.
Of these scientists of the later Middle Ages none has won greater fame
or survived longer in the popular mind than Albert von Bollstadt, known
both to his contemporaries and to posterity under the name of Albertus
Magnus (born about ixoo, died ii8o). He was of noble family, but from his
earliest youth devoted himself to learned studies and afterwards became a
member of the Dominican order, one of the then newly-founded orders of
mendicant friars. His reputation for learning spread rapidly throughout the
West; he was at one time a professor in Paris, being afterwards appointed to
a school in Cologne founded by the Dominicans, and finally becoming Bishop
of Regensburg. This last appointment, however, he did not hold for long;
he returned to the quiet monastic life and devoted himself entirely to science.
He believed his mission in life was to edit the writings of Aristotle — known
by him only in the above-mentioned Latin translation — and to harmonize
their results with the teaching of the Church. The majority of his many
writings deal with theology and philosophy, though natural science appears
to have occupied him most during the latter part of his life. As a natural
philosopher he is principally a chemist. He was the first to produce arsenic
in a free form and he made important discoveries in regard to particular
8o THE HISTORY OF BIOLOGY
combinations of metals; he also introduced into chemical terminology the
word "affinity" as denoting chemical relationship. As a biologist he follows
Aristotle, even where the latter's errors have been corrected by other ancient
philosophers; he holds that the arteries contain air, that the brain is humid
and cold, etc. The observations which he himself claims to have made are
often purely fantastical, but they also sometimes bear witness to his powers
of observation, which his chemical researches prove him to have possessed
in a high degree. His greatest service undoubtedly lies in his having directed
the world's attention to Aristotle's conception of nature and thereby also
indirectly evoking an interest in nature itself — an interest which the suc-
ceeding centuries were able to cherish and widen.
Contemporary with Albertus and, like him, a Dominican friar was
Thomas, called Cantimpratensis, after the monastery atCantimpre in France,
where he worked. His home was at Liege, but he studied at Cologne and
finally became a canon in the afore-mentioned monastery. His principal
work, De naturis rermn, forms, like that of his master, a compilation of the
nature theories of Aristotle and other classical authors, with a wealth of
notes on animals, both real and imaginary. More than Albertus, Thomas has
a penchant for weaving into his accounts of animals stories with a moral
point to them, and also, on the whole, he enters more into detail and is
less systematic than his master.
A third contemporary of these two, and a brother monk, was Vincen-
Tius Bellovacensis, who was likewise named after his monastery, at Beau-
vais in France. He wrote a work on nature entitled Speculum natura —
Nature's Mirror. This work is compiled from various sources: Aristotle in
Latin translation, Pliny, and Avicenna, as well as the Bible and the Church
Fathers. Though more haphazard and less lucidly arranged than those of
his colleagues mentioned above, it nevertheless had its influence on the age
and the succeeding centuries.
It is not worth while recounting further examples of this kind of medi-
aeval descriptions of nature — natural research it can scarcely be called. Those
already cited sufficiently show their character, that of a compilation of the
literary material of past ages in the service of that stock conservative theol-
ogy which dominated science during these centuries. But even at this period
there arose personalities whose ideas presage the intellectual liberation, the
foundations of which were laid in the course of these centuries and which
was destined later to overcome all obstacles during the Renaissance. One
such man was Roger Bacon (born 1114, died 12.94). By birth an Englishman,
he studied at Oxford and Paris and entered the Franciscan order, in which
he soon assumed a position of eminence. His liberal views, however, gained
for him bitter enemies, and once he was arrested and had to spend years in
prison, being deprived of every possibility of working until he was again
CLASSICAL ANTIQUITY, MIDDLE AGES 8l
released. No small cause of the mistrust he inspired was his interest in phys-
ical and chemical experiments, which resulted in his being suspected of
witchcraft and necromancy. Although he appears to have been a clever
experimentalist with a wide general knowledge, there is nevertheless no
record of any epoch-making scientific discovery that can be attributed to
him. His greatness lies in his general scientific ideas. He set himself up in
determined opposition to the subtle mode of thinking of the schoolmen and
urged that science should rather be based upon experience gained through
observing natural phenonema — that is to say, upon a method harmonizing
with that which has been adopted in natural-scientific research in more
recent times.
Appearance of neiv ideas
This intellectual emancipation from the hidebound teachings of authority
was inspired, however, less by the theoretical contributions of Roger Bacon
and any of his successors than by the increasing knowledge of nature itself
resulting from the discovery and exploration of new countries. The crusades
had already made some contribution towards this expansion, but still greater
was the influence of the knowledge of far-distant lands acquired through
the journeys into the interior of Asia undertaken by Marco Polo and several
of his contemporaries, and further through the widely extended voyages of
the Portuguese in the fifteenth century, and finally through the discovery
of America. As a result of all these geographical discoveries biology also ac-
quired a mass of fresh material, which it was impossible to deal with merely
by studying Aristotle; it forced research rather to seek its own paths and
research-workers to rely more upon themselves. Biology was thus compelled
to abandon the purely literary method of compilation and classification,
which had been the most characteristic feature of medixval science, and
instead had to rely for its progress upon working out its own observations
and developing the results thereof. But before it could do this, biology had
to free itself from the restrictions which the ecclesiastical authority of the
Middle Ages had laid upon man's intellectual activities in general, and it
thus came to take part in the great work of intellectual liberation whose
various phases in history are generally summarized under the name of the
Renaissance. The progress thus achieved in the knowledge of living nature
will be dealt with in the next chapters.
THE HISTORY OF
BIOLOGY DURING THE RENAISSANCE
CHAPTER XI
THE END OF MEDIEVAL SCIENCE
Revival of the study of ancient authors
THE UNIVERSAL SCIENCE of the Middle Ages, the philosophy of the
schoolmen, was, as has already been pointed out, a system of thought
complete of its kind, based on the infallible truth of the Catholic
Church doctrine, with a strictly formalistic conception of nature founded on
Aristotle. It was undoubtedly of service in its own time, especially in that it
developed the formal sides of thought, but it lacked the possibilities of free
expansion and it was thus inevitable that it should finally lose itself in bar-
ren subtleties. The intellectual movement which history calls the Renais-
sance was therefore hailed as a liberation of those in Europe who were true
seekers after knowledge. This movement started in Italy, where the connexion
with classical antiquity had never been entirely broken and where the system
of the mediaeval schoolmen had never really thrived; in the Italian colleges
during the Middle Ages Latin, rhetoric, and medicine were studied rather
than philosophy. The mediaeval Italian felt himself to be the rightful heir of
the old Roman people, and it was therefore natural that the cultural revival
in that country should take the form of a close study of ancient literature;
first of all it was the Roman writers of antiquity and later principally the
Greek authors unknown to the Middle Ages who here attracted the interest
which in other countries was devoted to the High-Church scholasticism and
who offered in exchange an entirely new and freer idea of existence than
mediaeval philosophy had been able to offer — an opportunity of developing
a more rich and many-sided human life than that which the Church of the
Middle Ages permitted. It was also in this sphere — that of the general
conception of life — that the great cultural revival in Italy exercised its
greatest influence, an influence of unique depth in spheres of culture, art and
literature, politics and economy. In the field of pure science this revolution
was, at least in the beginning, less complete; the absolute value of truth,
which the schoolmen ascribed to the formulas of the Church, the scientists
8i
RENAISSANCE 83
of the Renaissance, the Humanists, made over to the writers of antiquity.
Aristotle was regarded by them with, if possible, still greater respect than
by the mediaeval professors, the only difference being that now one had ac-
cess to the original writings of the master and could interpret them without
the restriction which the Church had formerly laid upon them. There was
no possibility, then, of any new conception of nature and its phenomena
developing in this direction. But fortunately there were other points of de-
parture for this development.
The fantastic speculations of neo-Platonism about infinity, and the al-
chemistic experimental science of the Arabs, formed the bases for a number
of attempts at an explanation of nature unfettered by Church dogmas and
scholastic systems, while, on the other hand, the great geographical dis-
coveries, as well as the newly-found classical authors, offered ideas for special
investigations in the sphere of biology which led to results far beyond those
of either Aristotle or Galen. The Renaissance period, therefore, was for the
science of biology a period of restless seeking and collecting, yielding results
which the succeeding age utilized for the purpose of making a complete
revaluation of the whole conception of nature common to the people of an-
tiquity and the Middle Ages. It would seem most convenient, then, first of
all to give a brief summary of the new philosophical speculations to which
the Renaissance gave rise, and then to examine in detail the results which
were achieved during that period by the science of biology.
CHAPTER XII
NEW COSMIC IDEAS AND NEW SCIENTIFIC METHOD
Opponents of scholasticism during the Middle Ages
EVEN DURING THE PERIOD whcn mediaeval scholasticism was at the zen-
ith of its power, there were not wanting movements hostile to it, the
representatives of which, partly by way of logic, like the so-called
nominalists, partly by exhortations to empirical observations, like Roger
Bacon, previously mentioned, sought to undermine its thought-structure.
These movements, besides, very often had points of contact with the mysti-
cism which throughout the Middle Ages sought in the sphere of a holy life
to induce a spirit of personal sincerity in contrast to the strictly formal piety
taught by the Church. When, later on, scholasticism was discredited, owing
to the reverence of humanism for antiquity, the field was left open for a
philosophy in which all the above-mentioned elements — theoretical specu-
lation, empirical observations, mysticism, both Christian and late classical
— were included as fundamental components in a fresh conception of exist-
ence, out of which our own modern ideas of nature and life gradually
developed.
The first important representative of this fresh view of nature was Nico-
LAUS CusANus. He took his name from the village of Kues, or Cusa, near
Trier, where he was born in 1401. He received his education amongst the
"brothers of the common life," a religious community having a pronounced
mystical tendency — an education which had a decisive influence on his
entire mode of thought. He had a brilliant career in the service of the Church,
into which he shortly afterwards entered. He became a bishop, and later on,
as a cardinal, he was one of the most trusted men of the papal supremacy at
the time. As such, he acted constantly in the interests of humanity and en-
lightenment, ardently opposing the sale of indulgences, trials of witches,
and other Church superstitions. He died in Italy in 1464.
New cosmic ideas
In the course of his manifold practical activities, however, Cusanus found
time for research work which places him in the first rank among the world's
pioneer spirits. The problems he deals with in his numerous writings are, it
is true, for the most part theological, but in connexion with them he touches
upon the problem of man's place in existence, and it is here that he makes his
most important contribution. Curiously interwoven with and derived from
84
RENAISSANCE 85
his mystical speculations appear his new and audacious ideas on the structure
of the universe and man's place therein. Basing his ideas on the mystical con-
ception of infinity of the neo-Platonists, he asserts that it is impossible for
the universe to have a spherical form, as Aristotle declares, for then it would
always be possible to conceive of something existing outside the sphere, in
which case it would not be the whole universe. Rather, the latter is infin-
ite; it exceeds all form and all limitations. Nor, in that case, can the globe
be the centre of the cosmos, for the cosmos has no centre, but man on the
earth imagines he is in the centre of the cosmos, and he would believe the
same were he to find himself on the sun or any other star. Cusanus thus
maintained the relativity of mental observation. He derives this knowledge
of his from what he calls " docta ignorantia (wise ignorance)," by which he
means the knowledge that all contrasts as well as all change in existence
finally become absorbed in an absolute maximum, infinite and unfathomable
as God Himself. It was this '' docta ignorantia" that Aristotle lacked, and
therefore he believed in a finite world and absolute mental observations. For
the rest, Cusanus employs his mode of thought ^uite as much in theologi-
cal sophistry, as, for instance, touching the true nature of the Trinity; but
while these subtleties are now long forgotten, through his ideas of nature
he takes his place as one of the pioneer thinkers of the beginning of the new
era, half medieval mystic, half modern natural philosopher. His bold ideas
seem otherwise to have attracted but little attention outside learned circles;
it was not realized how revolutionary they were, all the more so as he did
not concern himself with the details of our solar system; consequently he
did not attack the theory of the earth as the centre of the sun's orbit. His
high position in the Church undoubtedly saved him from such persecution
as afterwards befell many of those who deduced the inevitable consequences
of his theories.
If, then, Cusanus's ideas operated in silence, the views which about a
hundred years later were expressed by Copernicus attracted all the more at-
tention. Born in 1473 at Thorn in Poland, Nicolaus Copernicus received his
education at the Italian university of Bologna and finally became dean in
his own native city, where he died in 1543. Even in his youth he was a
keen student of mathematics and astronomy and already at that early age
began his life's work, to think out a new cosmic system which, more easily
than the Aristotelean-Ptolemaic, could be reconciled with the observations
made in his own time upon the movements of the heavenly bodies. Their
irregularities could in fact never be satisfactorily explained on the basis of
the old solar system. Copernicus discovered a better means of accounting for
the irregularities by letting the sun, in contrast to the direct evidence of the
senses, represent the centre of the cosmic system, and the earth assume the
place among the wandering planets which the sun held in the old system.
86 THE HISTORY OF BIOLOGY
Otherwise Copernicus retained that system practically unaltered; he thus
made the sun the immovable centre of the universe, the planets moving in
circles round it and the w^hole surrounded by the sphere of fixed stars, such
as the ancients imagined it. In reality, then, his theory was less subver-
sive than Cusanus's and, as a matter of fact, not without precedents in an-
tiquity; but still it attracted far more attention, because it was entirely at
variance with what everyone was accustomed to see happening daily before
his very eyes. Copernicus spent decades working out his theory, and not
until the year before his death did he dare to publish a book on it. It aroused
fierce opposition, particularly on religious grounds; the reformers as well as
the Jesuits condemned its teachings, while its scientific influence was at first
but small, all the more so as the proofs he offered of the truth of his new
theory were really rather weak. Shortly after his death, however, a thinker
was born who was able to reconcile Copernicus's ideas with those of Cusanus
and thereby founded a theory of the universe which in its essentials still holds
good today.
Giordano Bruno wa^born at Nola in south Italy in 1548. As a young
man he entered a monastery, but he was far from contented with the life
there, was soon suspected of heresy, and saved himself by flight. After this
he never found a permanent retreat; excommunicated and persecuted within
the Catholic world, he nevertheless found no consolation in the Protestant
countries which he visited. Upon returning to Italy he became a victim of
the Inquisition and after many years' imprisonment was condemned as a
heretic and burnt at the stake in 1600.
In numerous lectures, disputations, and published works he preached
in the countries he visited the new doctrine which cost him his life. In this
he takes as a starting-point Cusanus's speculations on infinity, Lucretius'
atomic theory, and Copernicus's solar system. On the ideas which he found
in these various conceptions he built still further with an originality which
ranks him amongst the greatest thinkers of all time, in spite of the fantasy
and mysticism with which he, like the other philosophers of the Renaissance,
burdens his speculations. In agreement with Cusanus, but still more emphati-
cally, Bruno maintains the subjectivity of mental observation; when a man
moves, the horizon goes with him, from which we must conclude that there
exists no absolute universal centre. On the contrary, both reason and faith
demand an infinite world, infinite as God Himself. And, like mental impres-
sions, place, movement, and time are relative and dependent upon the po-
sition in space from which they are observed. — Nor can the assertion main-
tained by Aristotle as to absolutely heavy and absolutely light bodies be
true; this being so, there is no meaning whatever in the old belief that planets
and fixed stars are lodged in spheres round the earth; on the contrary, they
move in their orbits in space freely and by internal force. And like his cosmic
RENAISSANCE 87
system Aristotle's theory of matter as potentiality in contrast to form
emanating from the divine intelligence is, in Bruno's opinion, also incorrect;
matter is rather the essential in every thing, the "divine essence" out of which
all is evolved. In Bruno's view, Lucretius' atomic theory and the neo-
Platonic ideas of the unity of matter are combined into a vision of the world
at the same time mysteriously vague and grandly fantastic, as a single whole,
one with God and one with itself, a combination of all the contrasts which
human thought has speculated upon. It would take too long to discuss these
ideas in detail, all the more so as Bruno, strictly speaking, does not touch
upon any purely biological problems. His importance from the point of view
of world history lies in the fact that he for the first time worked out, or
perhaps rather guessed, the cosmic theory which has since come to be held
in modern natural research. His influence has been great and has been widely
felt through all the ages.
While, then, in the cosmological sphere Bruno was the pioneer of the
new natural science, in a corresponding manner Francis Bacon (1561-1616)
paved the way in the sphere of pure laws of thought. His life's activities
and end were in all respects different from Bruno's. One thing, however,
they had in common: restlessness, that diversified seeking after knowledge
which was so characteristic of the Renaissance. Born in England in a refined
home. Bacon received a thorough education, but lost his father at an early
age and was not very successful in his official career, in spite of his brilliant
gifts and his ruthless ambition. It was not until later on in life that he re-
ceived higher appointments and eventually became Lord Chancellor in the
reign of James I, whom he knew how to flatter. But he was shortly after-
wards dismissed and condemned to pay a fine for bribery and corruption
when in office, and his last five years he spent in retirement.
Bacon's reform of science
Even in his early years Bacon had planned to reform all human knowledge
completely. This reform was to have been carried out in a work of mighty
proportions entitled Instauratio magna. Bacon during his restless life found no
time to carry out even approximately the great task he had set himself; the
"great reform" remained but a fragment, of which the first two, and the
best-constructed, sections are called The Advancement of Learning and The New
Method. The latter section is that on which Bacon's fame principally rests;
its title is chosen as a direct challenge to Aristotle, for whose theory of
method, Organum, Bacon wished to substitute his own new method. Bacon's
Novum Organum takes the form of a collection of aphorisms, intended to
illustrate from various points of view the inaccuracy of the traditional
scholastic mode of thought and the correctness of the new theory of thought,
which it was necessary to set in its place. The defects of the Aristotelean
philosophy are criticized in a number of strongly worded and merciless
88 THE HISTORY OF BIOLOGY
sentences; under its guidance human thought has been led into delusions,
which are classified under four different categories denoted by the names
"idols of the tribe," "of the cave," "of the market-place," and "of the
theatre." By idols of the tribe are meant those fallacies which are incident
to human nature, man's tendency to interpret the phenomena of nature ac-
cording to human preconceptions. The idols of the cave are man's individual
tendency to judge according to his own person or ego; it is as if a man sat in
a cave and from there saw things in a one-sided light. The idols of the mar-
ket-place are fallacies which arise out of human community-life, especially
errors arising from the influence exercised over the mind by mere words,
the confusing influence of traditional nomenclature upon the idea of things.
Finally, the idols of the theatre are those which are induced by the power of
tradition and result from received systems of philosophy and the tendency
of their theory to captivate the senses. The criticism which in further de-
veloping these principles he directs against the philosophy of his time is in
many cases extraordinarily sharp and holds good for every age. Thus he
utters insistent warnings against the common tendency to regard natural
phenomena as simple mechanical constructions, like those which man him-
self puts together and takes to pieces. Nature is, on the contrary, extremely
complicated and one must be. careful how one ascribes to its course of events
the same order and regularity which man loves to ordain for himself. From
this error arise fallacies such as the idea that the orbits of the heavenly
bodies must necessarily be circular, just because the circle is the most regular
figure. In contrast to the artificial and false idea of nature which the old
philosophy creates by means of such modes of thought. Bacon sets up the
true knowledge of nature, which is acquired by observation and experiment.
Man overcomes nature by obeying her laws and learns to understand her by
putting proper questions to her. Thus one arrives at the true scientific
method, that which by careful observation of the peculiarities of existence
and by a classification of them acquires knowledge of the general laws of
nature. Bacon attaches the very highest hopes to the value of the knowledge
of nature which he thus intended to create; throughout his long life he never
ceased to contemplate with passionate enthusiasm the thought of the ex-
traordinary life-values which awaited the human spirit in an enhanced
knowledge of the true essence of nature. Such knowledge could be attained
by means of a schematic procedure laid down once and for all, applicable
equally to high and low in the realm of thought. To this art of deduction,
however, based on a consideration of the temporal sequence of phenomena,
their presence and absence, and their numerical relations. Bacon gave a value
which it did not possess, and besides applied it in a manner which led to
sheer absurdities. His knowledge of nature, moreover, was limited and by
no means unprejudiced — he was, for instance, in opposition to Copernicus —
RENAISSANCE 89
and the experiments he arranged were primitive. Further, he was no mathe-
matician and he therefore was unable to employ the deductive reasoning
inherent in that science. Nevertheless Bacon has done an everlasting service
to the development of natural science, primarily through his activities as a
critic. It has already been mentioned how during the Renaissance ancient
culture in general and in the realm of science, primarily Aristotle, was treated
with unbounded respect. This uncritical and slavish attitude, which threat-
ened to ruin all chances of further progress, Bacon combats with all the
severity of which he is capable; he throws overboard all respect for antiquity,
whose culture he considers to have led only to intellectual decay and vain
disputes. He maintains that the peoples of antiquity were really children in
comparison with his own age, which possessed far more of that experience
which to him was the one foundation of knowledge. And in this insistence
upon experience as the sole source of knowledge lies his other great service
to the development of science and in particular to that of natural research.
He realized more clearly than any of his contemporaries the necessity of
extending the knowledge of nature by accumulating the results of obser-
vations of its objects and of experiments carried out with its powerful forces,
and though he himself could give expression to his ideas only in clumsy
efforts, nevertheless these ideas, through their intrinsic theoretical truth,
exercised great influence in his day and have done so down to the most recent
times. Even in our own day one of the pioneers of research into the problem
of heredity, Johannsen, has openly acknowledged his debt to Bacon's
Organum as the source from which he obtained a clearer idea as to the objects
and means of natural research. And it is certainly no mere accident that the
country which gave Bacon birth should have led the way in the great pioneer
work that has been done in promoting the development of biology.
What Bacon thus theoretically conceived and insisted upon was brought
to practical realization independently of him by Galileo, the creator of modern
physics and astronomy, and hence also the founder of the whole of modern
natural research and its conception of natural phenomena, so fundamentally
removed from Aristoteleanism.
Galileo Galilei was born in 1564 at Pisa, where his father held a good
post. At an early age he displayed mathematical and mechanical gifts, studied
at Pisa, first of all medicine and later mathematics, and when still quite
young was made professor of that science, first at Pisa and then at Padua.
In the latter city he worked as a teacher for eighteen years with brilliant
success; finally the university could find no hall large enough to seat all his
audience. And the results of his scientific work were still more brilliant;
especially after he had constructed a telescope and with it had begun to study
the heavenly bodies, his astronomical discoveries followed one another in
rapid succession: the globular form of the moon, the satellites of the planet
90 THE HISTORY OF BIOLOGY
Jupiter, the sun-spots, the phases of Venus and Mercury. But it was impossi-
ble to reconcile all these new facts with the ancient Aristotelean-Ptolemaic
cosmic theory and so Galileo early associated himself with the conception
of the universe as enunciated by Copernicus and Bruno. His great fame pro-
cured for him the personally brilliant appointment of Astronomer Royal to
the Medicean Grand Duke at Florence, with a high salary and no official
duties. But in leaving the service of the powerful Venetian Republic he came
under the influence of the power of the Roman Church, a circumstance all
the more dangerous to him as his new discoveries excited the bitter hostility
of the very parties which had condemned Bruno; moreover, he was himself a
violent controversialist, who never spared his enemies. His end is a matter
of common knowledge — how he was arraigned before the Inquisition on
account of a "dialogue" on the solar system and under threat of death was
compelled to make a public recantation of his "Copernican error," after
which he lived in strict seclusion until his death, in 1641.
Galileo' s theory
Galileo's fundamental importance as a natural philosopher is not based
merely upon his discoveries, epoch-making as they are; he has contributed
in a still higher degree towards scientific progress through the principles
which he laid down and which have become the basis of modern natural
philosophy. As we know, Aristotle based his cosmic theory upon the con-
trast between form and matter, where form is assumed to be a realization of
matter's powers of development; the higher the degree of its realization, the
more perfect the form. Therefore the heavenly bodies, with their regular
motions, are more form-perfect than the earth, with its many irregularities,
while beyond the heavenly spheres is the world of pure form, God, the origin
of all forms, the cause of all that happens in the universe. Galileo at once
came into conflict with this system through his astronomical discoveries;
according to Aristotle, the firmament, as existing nearest to the immutable
divine intelligence, was itself immutably regular in its motions. Galileo
discovered a great many irregularities; the sun-spots, Jupiter's moons, and all
else that the newly-invented telescope brought into the light of day proved
the firmament was not such a place of perfection and regularity as had been
supposed. On the other hand, the phenomena of motion in bodies here upon
earth showed an obedience to law of which the ancients had no notion.
Galileo experimented with the free f^ll of bodies, with pendulous motions,
and with the motions of bodies along an inclined plane, and discovered in all
these phenomena ratios between weights, lengths of time, and rapidity of
motion so mathematically regular that he could express them in the form of
theorems as capable of demonstration as the old geometrical propositions
formulated by Euclid. But it was just through this combination of natural-
scientific experiment and mathematical calculation that, as he himself says,
RENAISSANCE- 91
he created a new science. Instead of Aristotle's guiding reason, which was
in reality nothing but an expression for the speculating philosopher's own
inferences, deducted for the most part from purely human-cultural hypotheses
Galileo brings the phenomena of motion on the earth under one common
law, which operates out of mathematical necessity and whose manifestations
can under given conditions be calculated in advance, just as at an earlier
epoch it had been possible to calculate the regular path of the "divine"
heavenly bodies. Galileo was, it is true, unable to find one common law
governing the motions of terrestrial objects and the heavenly bodies — that
was for Newton to find in his law of gravitation — but Galileo laid down
the principle governing the natural-scientific treatment of terrestrial phe-
nomena, a principle which he expressed in the words: "To measure what can
be measured and to make measurable what cannot be measured." He seeks a
mechanical reason for everything that happens — a force that sets things
in motion. To refer to God as the cause of natural phenomena serves no pur-
pose, in his opinion, for one can attribute anything whatever to the will of
God, since no necessity underlies it. According to Galileo, natural science
should compare material things merely with one another, not with super-
natural things, and at the same time it has to be remembered that nature is
itself a miracle, although its phenomena have a natural explanation. In
actual fact, gravity is merely a word for something which we do not know;
we cannot tell what it is that attracts stones to the earth. Galileo sees clearly
that it is useless to try to find out what the forces of nature are; the scientist
can only discover how they operate.
Galileo' s victory over Aristoteleanism
Such a complete revolution of the aims and methods of natural science as
that carried out by Galileo could not of course penetrate men's minds all
at once. He himself fell a victim not only to the Church's intolerance, but
also to the superstitious respect in which the Renaissance held the culture
of antiquity and its chief scientific authority, Aristotle. Actually another
century was to pass before Aristoteleanism in every field of human knowledge
was successfully eradicated from the ideas underlying the science of the pres-
ent day. In order to rid natural science of Aristotelean fallacies it was, in
fact, necessary to destroy Aristotle's entire thought-system, and this was first
done during the seventeenth century by the great systematic thinkers of that
period, Descartes, Spinoza, and Leibniz, who will be discussed later on. We
shall now proceed to a survey of what the Renaissance period achieved in
the way of pure biological research, not only in the purely descriptive sphere,
but in the more speculative field as well.
CHAPTER XIII
DESCRIPTIVE BIOLOGICAL RESEARCH DURING THE RENAISSANCE
i'. Zoography
THE EARLIEST ACTIVITIES of Renaissancc research in the biological field
were, in accordance with the general tendency of that period, purely
philological. New editions of Aristotle, Hippocrates, Galen, and other
natural philosophers of antiquity were published, their language commented
upon, and attempts made to explain their contents. However, the actual
historical course of events compelled the learned world to carry out inde-
pendent work in regard to natural objects as well. Even the fauna and flora
of central Europe were very imperfectly known to the ancient philosophers,
and the information regarding many of them needed supplementing with
innumerable facts, which entailed much independent research work. And
this became all the more necessary when the great geographical discoveries
acquainted mankind with the perfectly new and exceptionally rich nature
of the tropics. All these circumstances combined to produce an abundant
literature of a purely descriptive kind, both zoological and botanical, which,
thanks to the art of book-printing, received such widespread publication as
the biological works of antiquity could never hope for. Further, the methods
of reproducing pictures, discovered in connexion with book-printing —
woodcuts and copperplate engraving — now for the first time made it pos-
sible to utilize the illustration in the service of scientific literature — a
means of extending human knowledge the importance of which can be ap-
preciated only if we consider what it means in our own day and what would
be the consequence if modern science were to be deprived of it. A review of
some of the more eminent representatives of this branch of biological science
during the Renaissance will give us some idea of the objects they aimed at
and the respects in which they advanced this science. For this purpose we
shall for the moment discuss only the results of zoological research during
this period; the botanical results may perhaps more suitably be left to a
subsequent chapter dealing with the history of biological classification.
Edward Wotton (i49z-i555) essentially represents the point of view
of mediaeval science. The son of a college porter in the University of Oxford,
he studied medicine in his native city and became a physician with a wide
and distinguished practice. His interest in nature he recorded in a lengthy
91
RENAISSANCE 93
work entitled De differentiis anhnalium, on which he worked for several
decades. In this book he shows himself a faithful follower of Aristotle,
whom he imitates both in his method of classification and in the field of
anatomy. His division of the animal kingdom is entirely Aristotelean: san-
guineous animals and non-sanguineous, viviparous and oviparous quadru-
peds, and so on. Nevertheless he criticizes his classical predecessors to the
extent that he does not accept without reservation the masses of fabulous
animals which they invent, but on the other hand he has nothing to say
about the many new animal forms which the explorers in his own century
brought home with them and which otherwise excited general interest
among his contemporaries, both educated and uneducated. Yet he contributes
much information regarding the medicines which may be extracted from the
various animal forms. As a profound exponent of Aristotle and representa-
tive of his ideas, Wotton came to exercise no small influence on his age,
particularly upon the man who eventually became the finest zoological
representative of the Renaissance, Gesner.
KoNRAD Gesner was born at Zurich in 15 16. His father was a Protestant
artisan, who fell in 1531 at the famous battle of Kappel, in which the civic
guard of Zurich, under the reformer Zwingli, were defeated by the Catholics.
Young Konrad, who had previously been sent to a good school, was now
unprotected, but his great reputation for zeal and energy brought him friends,
who sent him to study at their expense in Basel, Paris, and Montpellier. At
these places he studied such different subjects as classical and oriental lan-
guages, natural science, and medicine, and in general acquired in an unparal-
leled degree that many-sidedness in learning which during the Renaissance
was particularly appreciated and admired. After having been for some time
professor of Greek in Lausanne, he was appointed first town-physician at
Zurich, which was at that time a moderately salaried post. There he died of
a plague that ravaged the town in 1665 — that is, when he was still under
fifty. Of a quiet and unambitious nature, he had a constant struggle against
financial difficulties, which compelled him to wear out his strength in ill-
paid hack-work. His energy was marvellous. He published and made com-
mentaries on classical authors; he compiled dictionaries, wrote a lexicon of
classical literature, which must have been a very fine work for his period,
and was the author of works on popular medicine. Besides all this he found
time for extensive journeys both for scientific purposes and for pleasure —
he was one of the very first to be interested in mountain-climbing, and he
had a keen feeling for Alpine beauty — and finally he had the time and lei-
sure to carry out one of the greatest biological works the world has seen.
Gesner's Historia animalium comprises four immense folio volumes of
about 3,500 pages in all. The animals are arranged according to the principles
of Aristotle; the first part includes viviparous and oviparous quadrupeds, the
94 THE HISTORY OF BIOLOGY
second part birds, the third fishes, the fourth, which was published after the
author's death, reptiles and insects. In each part he then describes one animal
after the other on the lines of Pliny, but with far greater expert knowledge,
based oa his own experience and criticism of his source of information.
Animals are arranged alphabetically "in order to facilitate the use of the
work," though allied forms are grouped under one heading; all oxen under
Bos, all apes under Sitnia, etc. Each animal form is discussed under eight sec-
tions, marked with letters of the alphabet and comprising (a) the name of
the animal in different languages; (b) its habitat and origin and a description
of its external and internal parts; (c) "the natural function of the body";
(d)the qualities of the soul ;(e) the animal's use to man in general ;(f) its util-
ity as an article of food; (g) its utility for medical purposes; and (h) poetical
and philosophical speculations about the animal, anecdotes and resemblances
to be found in different authors. Thus the reader is able to find what he wants,
whichever part of the work he turns to. This clearly shows its encyclopasdic
character, and actually it is far more reminiscent of Pliny than of Aristotle.
As in Pliny, so in Gesner one seeks in vain for any idea as to the connexion
in living nature, in vain for any comparison worked out between the differ-
ent forms of life, regarding their organs or their functions. Gesner, however,
surpasses Pliny in knowledge — in this respect, of course, he has the whole
of the intermediate literature at his disposal, and indeed he has it at his
finger-ends. True, he, too, brings in a great many stories of marvellous
animals, but he certainly has not that absolutely unquestioning belief in
the miraculous which the old Roman had. And, above all, he was able to
record the results of his own research work, for he studied not only books,
but also life. He was a keen collector of observations on animals, not only
his own, but also those of other scientists with whom he corresponded.
Illustrations introduced into xpology
His most original contribution to science was his introduction of the illus-
tration as an aid to the study of biology. He desired that, as far as possible,
every description of an animal should be followed by an illustration so as to
give the reader a clearer idea of the animal, and he spared neither trouble
nor expense in procuring as accurate woodcuts as possible. His collaborators
in this work were eminent artists, and he himself declared that the picture
of the rhinoceros was done by no less a person than Albrecht Diirer. With all
its weak points, Gesner's Historia animalium is at any rate the foremost
purely zoological work of the Renaissance period, and its influence on the
science of the succeeding age was considerable.
Somewhat younger than Gesner and partly his pupil was another of the
foremost zoologists of the Renaissance, Ulisse Aldrovandi. He was born
at Bologna in 15x1 of a respectable burgher family and was intended to be a
merchant. Office work, however, attracted him but little, and so he went in
RENAISSANCE 95
for studying, first jurisprudence in his native town and then philosophy and
medicine at Padua and Rome. When he was thirty years old, he took the
degree of doctor of medicine and shortly afterwards, in 1560, he was made
professor at Bologna, where he worked for forty years, resigning at the age
of nearly eighty. He died in 1605. As a professor he lectured chiefly on
pharmacology, and to aid him in his teaching he planted a botanical garden.
This caused him to fall foul of the apothecaries of Bologna, who alleged
that in cultivating medicinal plants he usurped their privileges. The contro-
versy grew so fierce that it finally had to be settled by the Pope. Aldrovandi
was, on the whole, a man who lived for his science; he spent his fortune on
collecting natural objects and had recourse to the leading artists of his time
to draw pictures of them. The Government of Bologna doubled his salary
in recognition of his great services to science, and in return he bequeathed
to the city his collections and library.
Aldrovandi' s natural history
In his energy and capacity for work Aldrovandi resembled Gesner, and as
he lived longer and worked under more favourable conditions, he managed
to achieve far more. His collected works on natural history consist of four-
teen large folio volumes, besides which there are preserved in the University
of Bologna quantities of unpublished manuscripts in his own handwriting.
He himself published during his lifetime only four volumes, on birds; after
his death his friends and pupils published the remainder: those on other
animal groups, on plants, and on stones. These latter volumes, however,
seem to have been in part radically revised by the editors, wherefore Aldro-
vandi should be judged only on what he himself published. His model was
chiefly Gesner, whose work he diligently studied and it is from this point of
view that his own work must be judged. His relation to Gesner is by no
means in every respect that of an improver; he is far less critical, and similarly
he has on the whole less stylistic ability; in his descriptions he piles up
masses of like and unlike, so that one of his most eminent successors, BufFon,
was moved to express the opinion that only one-tenth of the whole of Aldro-
vandi's works would be left if one extracted all that was useless and untrue
from his writings. On the other hand, his illustrations, as well as his typo-
graphical equipment, are better than Gesner' s, while, at least in some re-
spects, he is in advance of the latter in regard to classification. Birds are
classified according to certain groups: first, birds of prey; then wild and tame
fowl (gallinaceous birds) — characterized as " pulveratrices" ; i.e., those that
bathe in sand — further, pigeons and sparrows, which bathe in both water
and sand; then song-birds, baccivorous and insectivorous; and lastly water-
fowl. Moreover, he has paid attention to anatomy, particularly osteology;
and, finally, he cites a far greater number of exotic and hitherto unknown
forms than Gesner. He too, then, has in his own degree contributed to the
96 THE HISTORY OF BIOLOGY
advance of biology, and though he by no means merits the vaunting eulogy
which a contemporary artist wrote under his portrait — that, though not in
his appearance, yet in his genius he resembled Aristotle — nevertheless his
work has exercised a powerful influence, and it was not until BufFon's great
zoological work in the eighteenth century that Aldrovandi's was definitely
out-distanced.
Apart from these representatives of the knowledge of the animal world
as then known, certain research-workers are worthy of mention who devoted
themselves to the study of particular animal groups with which they dealt
monographically. In the best of these monographs there is really far more
evidence of independence in research and originality of ideas than in the great
collective works; in the former is best seen that power of independent obser-
vation and investigation of natural objects which was a feature of the science
of the Renaissance.
GuiLLAUME RoNDELET was bom in 1507 at Montpellier in the south of
France, where he also worked later as a professor. He studied first in his own
district and then as body-physician to a distinguished personage on his travels
in Italy, where among other people he made the acquaintance of the young
Aldrovandi, who received much sound teaching from him. As a professor he
established in his native city an anatomical theatre, but he had not been
working there long as a teacher when he died, in the year 1556. His fame as
a biologist rests on his work De fiscibus marinis. In this book he describes
and illustrates the aquatic animals he knows, for he regards as fishes not
only seals and whales, but also crustaceans, molluscs, echinoderms, worms,
and other marine invertebrates. He makes a very careful study of whales,
fishes and cephalopods. According to his own statement, he dissected a large
number of these creatures and he also gives a number of correct particulars
which are sometimes at variance with the great authority Aristotle. He fur-
ther compares, as far as was possible, the same organs in different fishes, giv-
ing exact accounts of different maxillary and dental forms, different branchice,
etc. However, his comparative work practically gets stranded, owing to the
impossibility of finding resemblances between the vertebrate and invertebrate
forms discussed. His attempts at classification are likewise very primitive.
He differentiates between selachians and osseans, which again are divided
into "flat" and "high" fish; moreover, whales are dealt with in a group by
themselves. He has as little notion of species in the modern sense as had
Gesner and Aldrovandi, and therefore, like them, he had to begin the de-
scription of every form by recounting as many names for it as possible. On
the other hand, he avoids for the most part the useless petty details of scholar-
ship with which the two last-mentioned authors of collective works over-
burdened their accounts, and this at once gives to his work an impression of
greater accuracy. And though he certainly does illustrate a number of mar-
RENAISSANCE 97
vellous creatures reported to have been seen, such as, for instance, a fish hav-
ing "the appearance of a bishop," he does so w^ith a reservation as to the
irrational nature of stories relating to such phenomena.
Besides Rondelet, a younger countryman of his is worthy of mention,
Pierre Belon, in several respects a man possessing great ideas about the
future. He was born in 15 17 near Le Mans, in central France, of poor parents.
His genius attracted the attention of the bishop there, who defrayed the cost
of his medical studies in Paris. After that Belon went to Germany for a course
of study. Upon returning to France he received, through the kindness of
certain distinguished patrons, funds to enable him to undertake a still
longer journey, through Greece, Turkey, Syria, and Egypt. Everywhere he
collected material with great energy and made notes, not only on natural-
scientific, but also on archaeological and ethnographical subjects. On his
return home he settled in Paris, where he was granted a pension by King
Henry II. In 1564 he was murdered by highwaymen. His period of scientific
authorship was thus brief, lasting not more than about ten years, but during
that time he brought out ideas of great significance for the future. He was
held in high esteem even by his contemporaries; he counted amongst his
friends the famous poet Ronsard, who wrote verses in his honour.
Like Rondelet, Belon devoted himself to the study of marine animals and
published two monographs: UHistoire naturelle des estranges poissons marins
and La Nature et diversites des poissons. The term "fishes" he makes even more
comprehensive than Rondelet: not only whales and seals, crustaceans, mol-
luscs, and actinic-e, but the hippopotamus, the beaver, and the otter are also
described amongst the fishes. And even if all these animals could be classified
by a faithful Catholic as fishes, just because the Church included them among
the animals that may be eaten during a fast, it is hard to understand why the
chameleon and the uromastix lizard are catalogued in the book — these
beasts of the desert which have nothing whatever to do with water. Though
the external grouping of the subject, then, leaves much to be desired, Belon
has certainly endeavoured to introduce into the class of true fishes some kind
of systematic division, based not merely on external, but also on internal
anatomical characteristics. Cartilaginous and bony skeletons, Ovipara and
Vivipara, constitute the bases of classification, which still hold good today,
and on the whole his system of classification bears a somewhat more modern
stamp than Rondelet's. Even attempts at an investigation of various forms
on the lines of comparative anatomy occur in Belon's work. Whether and,
if so, to what extent he was influenced by his immediate predecessor it is
difficult to decide. Their works were published practically at the same time.
As regards wealth of material, at any rate, Belon's work is superior. More-
over, thanks to his travels, he was able to include many oriental animal
forms which were previously unknown to the Western world.
98 THE HISTORY OF BIOLOGY
Far superior to the works on the fishes is Belon's second main treatise:
Histoire des oyseaux. In this work he describes and illustrates all the birds he
knows, arranged in groups according to their structure and habits — birds
of prey, waterfowl, shore-birds, ground-pecking, wood-pecking, omnivo-
rous, and small birds, divided into Insectivora and Granivora. The individual
forms are characterized by a few names in Latin, Greek, and French; unlike
Gesner, Belon scorns to extend his knowledge of languages. If this attempt
at classification bears witness to Belon's keen powers of observation, there
is still further proof of them in the attention he pays to the morphology and
anatomy of the individual forms. The structure of beaks and claws is closely
studied and compared in different forms, while anatomical relations are
treated in the same way. Most noteworthy, however, is the detailed com-
parison in both text and illustration, in the first book of the work, between
the skeleton of a man and that of a bird, the latter drawn in an attitude cor-
responding to that which the former assumes in his natural standing position.
Although this comparison by no means agrees in every detail — for instance,
the human clavicle and the bird's coracoid bone are made homologous — at
any rate we have here a first attempt at a comparative anatomical investi-
gation. The idea thus started by Belon was, it is true, for a long time neg-
lected; it was not until two centuries later that it was taken up anew by
Buffon, to be eventually developed by Cuvier into one of the most important
fields of biological research. The fact that these two were both countrymen
of Belon is indeed some evidence that his activities in this sphere did not
pass entirely unnoticed.
z. Anatomy
That the age of comparative anatomy had not yet arrived was of course due
to the fact that research was still fully occupied with purely descriptive
anatomy. In this, as in other spheres, the Renaissance inherited from the
great anatomists of antiquity, of whom Galen constituted the chief author-
ity, in his reputation comparable with Aristotle, and like him regarded with
infinite respect by the physicians of the Renaissance, who were philologi-
cally rather than biologically educated. However, it was not the phy-
sicians alone who required anatomical knowledge; even art, the result of
the admiration of the Renaissance for antiquity, now began to demand a
closer study of the structure of the human body. Amongst the pioneers in
this field the first name that should be mentioned is that of the great universal
genius Leonardo da Vinci (i45x-i5I9).
Leonardo was a Florentine, and in his native city, which was the very
centre of Renaissance culture, he was brought up to be an artist and at the
RENAISSANCE 99
same time a mechanician — the professions of painting and mechanics in
those days were often combined. He afterwards led a restless life, carried out
work in many places in Italy, and ended his days at the French Court. His
world-wide fame he of course won as a pioneer in the art of painting. In
this his greatest contribution was his introduction of a close study of human
anatomy; he drew for the benefit of his pupils a vast number of anatomical
figures, which are still preserved, and published a work on the proportions
of the human body. He did not study only man, however; all sorts of natural
objects and natural phenomena interested him. In a mass of roughly drafted
notes, which were never combined into a connected whole and were not
printed until our own day, he has recorded his observations and reflections
on practically every sphere of human knowledge. He not only studied human
anatomy; he also compared similar organs in different living creatures; he
investigated optical sensations; he observed the structure of different geo-
logical strata, and maintained, in opposition to Aristotle, but in agreement
with Xenophanes, that fossils were animal remains. In every field he shows
himself an opponent, not only of scholastic traditions, but also of the slavish
admiration for antiquity that characterized the Renaissance. Not the classical
authors, but experience should be the source of human knowledge. Unfor-
tunately his speculations were merely fragmentary and for that reason were
unprinted. It was not until later that they were more closely studied, and
Leonardo's influence upon science has been only indirect, as a result of the
impression made by his personality and his art. The study of the human
anatomy which he initiated was continued by other Renaissance artists and
in that way reacted upon culture in general; these painters and sculptors
were certainly not without their influence upon the impetus given to medical
anatomy in the sixteenth century.
In the field of medical science the influence of the Renaissance was felt
in the same way as in other branches of human knowledge; a return was made
from the mediasval authorities to antiquity. The doctors applied themselves
to the study of the classical languages; they severely condemned the barba-
rous Latin of the mediaeval professors and formed their style on the best
Roman and Greek models. And in conformity with this enlightened spirit
the poor editions of the medical writers of antiquity which were based on
Arabic translations were banned and were replaced by new editions of Hip-
pocrates, Celsus, and Galen, which, published with careful textual criticism
and sound commentaries, were, thanks to the progress of book-printing,
widely dispersed throughout the universities. One of the most brilliant and
at the same time most typical students of medicine during the Renaissance
was Jacob Sylvius, of Paris. Born in 1478, he devoted himself from early
youth to the study of classical languages, not only Roman and Greek, but
also Hebrew. He was a fine stylist and was the author of several works on
lOO THE HISTORY OF BIOLOGY
French grammar. He was nearly fifty years old when he took up medicine,
applying himself to the study and exposition of classical medical literature.
In lectures, which were brilliant in their formal delivery, he expounded to
the students of Paris the theories of Galen, which to his mind were infalli-
ble — "divinely inspired" — and could not be improved upon. These lec-
tures were really exercises in classical oratory; there was no question of any
empirical research. Practical instruction remained in all respects at the point
to which the Middle Ages had advanced it. And in this respect the Middle
Ages did actually advance beyond antiquity.
Mediaval dissection
As early as the middle of the thirteenth century dissections had to be
carried out on human bodies at the Italian universities; the Emperor Fred-
erick II, who had no prejudices, made it compulsory for students of medicine
and surgery to attend these operations, while later on, the prohibition of
the popes was powerless to prevent the development of these practices. At
the universities of Salerno, Bologna, and Padua they were officially ordained
and had, if possible, to be carried out regularly. As a result of this it should
have been possible to leave the work of Galen at its worth, for, as we know,
he had never dissected human bodies, and his anatomical dicta were very
unreliable and highly misleading. This, however, was not to be; the Middle
Ages were far too bound by respect for authority, and in particular the classi-
cal authorities. Moreover, the study of anatomy was rendered difficult on
account of the antagonism prevailing between the physicians and the sur-
geons. Members of the faculty of medicine pursued their studies only on a
literary and speculative basis and looked down upon the surgeons as merely
a body of artisans. In dissections it was always a surgeon who wielded the
knife, while the professor, staff in hand, pointed out and demonstrated what
was brought to light. The results of this collaboration were also primitive.
The surgeon's instruments and hold were the simplest imaginable; with a
knife — the use of a saw and chisel, probe and canula was unknown then —
the abdomen and chest cavity were opened and the internal organs laid bare
for examination. After this the idea was that muscles, nerves, and blood-
vessels should be exposed for study, but this was usually too difficult a task
for the operators, nor did it amuse the students, who very soon marched off
unless the proceedings ended with one of the professorial discussions that
were so popular at that time. The professors of the faculty of philosophy,
who were usually invited to attend the proceedings, taking Aristotle as
their authority, would attack Galen, who would be courageously defended
by all the medical professors present. The differences of opinion between
these great authorities of the ancient world could be dragged out into end-
less discussions and give rise to the most absurd sophistical arguments. This
was the course that anatomical studies were still taking in the sixteenth
RENAISSANCE lOI
century, so that nothing was to be expected from them in the way of biologi-
cal development. Then a man came upon the scene who at once led anatomi-
cal research into a completely new direction, created an entirely original
method of procedure, and thus started a new era in the history of science.
Andreas Vesalius was born in 15 14 or 15 15 at Brussels, of a family
which had taken its name from the district of Wesel in the Rhine Province and
which for several generations had been devoted to the medical profession. He
himself chose the same profession and prepared himself for it by a thorough
school-education. Though his studies were exclusively humanistic, for there
was no other kind of education given in the schools of those times, the young
Vesalius was able in his own way to satisfy his craving for biological knowl-
edge; he studied ancient anatomical works which he found in the family
library and himself dissected animals of various kinds which he managed to
procure. At the age of eighteen he went to Paris in order to study medicine
seriously. There, however, Sylvius, whom we have mentioned above, was the
ruling spirit, with his classical-philological method of education. Vesalius
had again to rely upon his own resources. And his force of will enabled him
to make a way for himself. He began to collect bones from the places of
execution and went on with his dissection of animals. Soon he acquired such
a reputation that he was called upon by physicians and students to perform
public dissecting operations on human bodies in place of the surgeon, and he
fulfilled his task so well that not only the internal organs of the corpse, bui;
also the muscles, nerves, blood-vessels, and bones were completely demon-
strated. After three years he left Paris, worked at his home for a brief period —
in the course of which he succeeded, inter alia, in putting together a complete
skeleton out of bones from the gallows — and then went to Italy. In Venice,
where there existed at that time a very keen interest in medicine, he increased
both his learning and his reputation, with the result that, immediately after
he had graduated, he was appointed professor at Padua, at the age of twenty-
two, after only four years of study. He could hardly have wished for a more
satisfactory field of activities. An enlightened government, an interested
audience, and a thoroughly educated public all equally favoured the attain-
ment of his ambitions. And, indeed, Vesalius surpassed all expectations. His
interest in his science was indefatigable and his enthusiasm for imparting
his knowledge inexhaustible. His demonstrations on dissection used to
bring together as many as five hundred listeners, and this in spite of what,
according to the ideas of his time, were unheard-of claims that he put upon
his audience, which he kept busy from morning to night for a space of three
weeks. Dissection lectures, which were always held in the winter so that
the material should not putrefy, began with a demonstration of the skeleton,
the bones of which were carefully gone over; then the muscles, blood-vessels,
and nerves of one corpse were prepared, and finally the internal organs of the
lOl. THE HISTORY OF BIOLOGY
abdomen and chest of another body, as well as the brain. Vesalius himself
used to perform the essential work in dissecting, assisted by students; the
surgeons, who elsewhere had such an important part to play, had nothing
whatever to do here. A mass of new surgical instruments came into use; the
models had been partly invented by Vesalius himself and partly borrowed
from among the tools owned by a number of artisans whom he visited in
order to initiate himself into their technical ideas.
Vesalius' s great anatomical ivork
In his demonstrations Vesalius was at first a faithful follower of Galen, for
whom since his youth he had entertained the greatest respect. It soon became
more and more obvious, however, that Galen's observations were incomplete
and that his presentation of them was vague and self-contradictory. The
more zealously Vesalius anatomized, the more did he realize how necessary
it was to reproduce in print all his anatomical observations and, without
reference to any authorities, to describe the structure of the human body as
it really is. The result was his two literary masterpieces: De humani corf oris
jabrica, a large folio volume of over seven hundred pages, and a compendium
of the same, Epitome, of thirty-one pages, both containing numerous illus-
trations by eminent artists after Vesalius's original preparations. These books
were published in 1543 at Basel, where Vesalius spent a whole year's leave
of absence superintending the printing. Through these two works Vesalius
created the modern science of anatomy. They made an enormous impression
on his contemporaries. Galen's followers were furious, particularly Vesalius's
old master Sylvius. Many were the polemical treatises written in opposition
to the man of dangerous newfangled ideas, and the rage of his opponents can
still be perceived from the controversial methods they adopted. They were
not content with merely declaring that Vesalius's work was absolutely
inferior; the most abominable and absurd accusations were heaped upon his
personal character — he was godless, he was sordid; like the ancient Alex-
andrian anatomists he had dissected men alive (sentimentality as regards
animals was not so deep in those days as to make it worth while quoting
the vivisection he actually performed on animals). Even after his death his
memory was treated with contumely, especially in France, where the fol-
lowers of Galen were to hold unrestricted sway for another hundred years.
Vesalius could now no longer hope to enjoy the tranquil conditions under
which he had worked in the past. In the year after the publication of his
writings we find him relinquishing his professorship and accepting the post
of court physician to the Emperor Charles V. What induced him to take this
step is not known; it is assumed that after the completion of his anatomical
masterpiece he wished to devote himself to practical medicine, but that he
might just as well have done in Padua. It is more likely that he hoped that
in his appointment at the court of the most powerful monarch in the world
RENAISSANCE 103
he would find protection from the persecution of his enemies. Besides, several
of his ancestors had been court physicians. He accompanied his delicate
master on all his many journeys through various European countries, in the
course of which he had but little time for continued research. In 1555, how-
ever, he published a new and improved edition of his great work, in which
he vigorously refutes his calumniators. Upon Charles's abdication he joined
his son Philip II, whose notorious obscurantism offered but the smallest
chance for his personal retainers to develop liberal ideas. In fact, after eight
years we find Vesalius leaving the court; in 1564 he visited Venice, in the
hope apparently of again taking up his old professorship, which then hap-
pened to be vacant. While waiting to be appointed he made a journey to the
East, visited Jerusalem as a pilgrim, and never again returned to the West.
The reason for his journey is not known for certain, nor indeed how he ended
his life. Thus disappeared into the unknown one of the greatest scientists of
modern times.
Vesalius 's great anatomical work is arranged in the same order as that
which he followed in his anatomizing, mentioned above: first he discusses
bone-construction, then muscles, blood-vessels, and nerves, then the ab-
dominal organs and those of the thorax, and the brain, and finally he devotes
one chapter to an account of his vivisectional method. In his general concep-
tions Vesalius entirely adopts the standpoint of antiquity. His division of the
component parts of the body into simple and complex is borrowed from Aris-
totle, as also are most of his physiological terms; the food is "cooked" in the
abdominal cavity, the object of respiration is to cool the blood, the embryo
arises out of the father's semen and the mother's menstrual blood. From Galen,
whom he still highly respected, he takes his general conception of continuity
in existence and of the causes that govern it. The Creator has, to His own
honour and to the benefit of man, made the human body as perfect as pos-
sible; every part of it has been created just as it is in order to fulfil its specific
purpose. In many important details also he adheres to Galen's ideas, particu-
larly in regard to the circulatory system; he gives, it is true, an exhaustive
description of the structure of the heart, but as regards its and the liver's
relation to the vascular system he still retains the old traditional view. The
greatness of Vesalius lies in his method and technique. In this he created
the conditions necessary for the development of modern anatomy. Most of
the technique which is practised in every anatomical theatre today originates
from him; the instruments used at the present time are practically the same
as those which he designed, and the majority of them he introduced into
dissectional practice; the course of instruction as laid down in his works is
still followed; the skeletons used for demonstration purposes are mounted
after his method; and the plates used to facilitate instruction are for the most
part merely improved editions of his own. But his great service to science is
I04 THE HISTORY OF BIOLOGY
not confined to this. In almost every sphere of human anatomy he made im-
portant discoveries in matters of detail and, still more, corrected old fallacies.
To enumerate all that he did in this respect would be impossible here; but he
who cares to do so can compare, for instance, a picture of a skeleton made by
any one of his predecessors with one of his. If we add to this his masterly
and at the same time exact and highly imaginative descriptive method, with
its splendid simplicity of arrangement in the midst of a wealth of detail,
every impartial judge must admit that he was, in spite of his lack of original
ideas, one of the greatest biologists that have ever lived.
Vesalius's influence on the development of anatomy was primarily of
benefit to Italy. In France the disciples of Galen still upheld the authority
of their master; in Germany the interest in the study of nature was being
gradually suppressed by interminable religious strife. Italy, on the other
hand, thanks to the impetus given by Vesalius's practice of anatomy in
Padua, became throughout the succeeding century the centre of anatomical
study. Vesalius's pupils followed in their master's footsteps and carried on
his work by widening the field of detailed research. His prosector and suc-
cessor in Padua, Realdo Columbus (date of birth unknown, died 1559}, made
a special study of the organs of hearing and the blood-vessels in the lungs.
He published the results of his experiments in a work entitled De re anafomica,
in which he shows himself a well-informed anatomist, but a not very sym-
pathetic personality, self-opinionated and overbearing, not least towards
his old master. He was soon called away, however, to carry on other activi-
ties elsewhere and was succeeded in Padua by a man of far higher qualities,
Gabriele Fallopio. Born in 152.3, Fallopio spent his youth in poverty, was
for a time in the service of the Church, but afterwards had an opportunity
of studying anatomy in Padua, probably during the very last years of Vesa-
lius's professorship. His career was as rapid as the latter's; at the age of
twenty-four he was a professor in Ferrara, whence he was summoned to
Padua, where the Government maintained him in every way. He carried on
the Vesalian traditions with honour, attracting to his lectures a large
audience and at the same time working at an extensive medical practice.
Unfortunately his life was short; he died in his fortieth year. During his
lifetime he published only one, rather small, but useful, work entitled
Observationes anatomka. In its introduction he speaks most highly of his mas-
ter, Vesalius, and with the greatest modesty of his own observations. These
are, however, in certain respects of fundamental importance. In particular,
he increased the knowledge of the sexual organs — in this field the Fallopian
tube bears his name — while his contribution to the knowledge of the
structure of bone and of the organ of hearing was of considerable value.
But he also made important discoveries in most other fields of human
anatomy. Besides this his activities extended to other spheres of medical
RENAISSANCE 10$
science; the results he achieved here were not published until after his
death.
To Fallopio's professorship, which, as mentioned above, Vesalius had
hoped to resume, was appointed after the latter's death another scientist
who was also a pioneer in his branch — Girolamo Fabrizio, usually called,
after the place of his birth, Fabricius ab Aquapendente, to distinguish him
from a contemporary German anatomist Fabricius. Born in 1537, he studied
under Fallopio, was his prosector, and succeeded him as professor in 1565.
In contrast to his famous predecessors he lived to a good old age: he died in
1 61 9, having been for ten years emeritus professor. Besides anatomy he lec-
tured on surgery; he raised that despised "handicraft" to the rank of a
science and was himself an eminent practitioner, his profession bringing him
immense wealth, which he generously utilized for the benefit of science. Ana-
tomical research was in his time liberally patronized by the Venetian Govern-
ment, which built a fine anatomical theatre and paid generous salaries to its
staff.
Fabrizio was a very productive scientist, though more qualitatively
than quantitatively. His predecessors had devoted themselves exclusively
to human anatomy, and such contributions to comparative anatomical re-
search as had been made in other quarters — Pierre Belon's, for instance —
had passed practically unnoticed. Fabrizio adopted the method of compara-
tive research, which really no one since Aristotle had applied with anything
like original results, and he developed it further in one of the most important
spheres of biology — namely, embryology. His treatises on the evolution
of the egg and the embryo present in clear and concise form, with good illus-
trations, the process of embryonic development in a large number of verte-
brates: birds and reptiles, mammals and sharks. He describes the anatomy
of the embryo and the shape and appearance of the placenta and embryonic
tissues, pointing out the similarities and differences between the various
animal forms, with a wealth of hitherto unknown facts, which it would take
too long to follow in detail. Fabrizio employs the same comparative method
in a number of other spheres of biology. Thus, he describes the movements
of animals from a comparative point of view; again he studies the noises of
animals. This leads him to make an interesting attempt at animal psy-
chology — certainly the first of its kind. Amongst his purely anatomical
works may be noted his investigations into the structure of the ear, the eye,
and the larynx. Of more definite value to posterity, however, was a three-
page article on the venous valves, which he discovered experimentally —
through binding the limbs of live human subjects for the purpose of bleed-
ing — and which he afterwards closely studied with reference to their
structure and distribution. In spite of this discovery, which was so obviously
at variance with the Galenian theory of circulation, he could not abandon
lo6 THE HISTORY OF BIOLOGY
the latter; he explained away his discovery, and it was left to one of his
pupils, the Englishman Harvey, using this fact as a starting-point, to formu-
late a true conception of the circulation of the blood.
Fabrizio was the last of the great anatomists of Padua — a line of great
men in the service of biological research, such as scarcely any other university
has been able to produce in an unbroken sequence. But besides these Italy
possessed in the sixteenth century a great number of eminent specialists in
the field of anatomy. Space, however, does not permit of our dealing with
more than one or two of them as examples of the enthusiasm with which
anatomical research was carried on in the country in which Vesalius stimu-
lated such interest in that branch.
Bartolommeo Eustacchi was a student of research possessing wide in-
terests and deep knowledge, which, however, owing to the unhappy fate
that befell his works, came to have but little influence on the progress of
science. The date of his birth and the early circumstances of his life are un-
known to us; in the middle of the sixteenth century we find him in practice as
a physician in Rome and then as professor at a papal medical academy. He
died in 1574. He had recorded his widely extensive anatomical investigations
in a richly illustrated work, which at his death was ready for the press. It
was withdrawn, however, and never published until 1714, when most of it
was naturally out of date. During his life, Eustacchi found time to publish
a number of smaller treatises, Opuscula anatomka, among which were several
important investigations, as, for instance, that of the auditory organ, in
which the Eustachian tube still bears his name, and of the blood-circulation
and dental development in the embryo.
Another eminent anatomist was Costanzo Varolio, of Bologna (1543-
75), who in the course of a short life managed to carry out important in-
vestigations into the nervous system, in which the pns Varolii in the brain
is named after him.
Far more remarkable than these two, however, is Cesalpino, a scientist
who made weighty contributions in several different fields of research; in
biology, as a speculative natural philosopher, and as a botanist. His life's
work, however, is best described in another connexion, among the pioneers
in the discovery of the circulation of the blood.
The position of Marc' Aurelio Severino among the Italian anatomists
is a curious one. Born in 1 5 80 in south Italy, he came at an early age to Naples,
where he studied the humanistic sciences and philosophy under the famous
Campanella, known as a keen opponent of Aristotle and as a victim of
political and scientific persecution. Soon, however, Severino began to devote
himself to the study of medicine and was appointed professor of anatomy
and surgery at Naples. He had, besides, a wide medical practice. At one time
he was subjected to persecution by the Inquisition and had to flee from Naples,
RENAISSANCE 107
but he was soon recalled and throughout his life enjoyed a great reputation.
He died in 1656 of a serious plague, which he had endeavoured to stamp out.
He wrote a handbook of human anatomy, a monograph on the viper, and,
finally, the work which made his name famous: Zootomia Democritea.
Zootomia Democritea
Severing introduces his work with a defence of the comparative study of
the anatomy of different animals, the advantages of which he demonstrates
in a dichotomously arranged table, and then further dilates upon them with
a mass of quotations from other authors and arguments of his own. He finds
its best to begin the study of anatomy with animals, as they often have a
simpler and more easily accessible organization than man, with whom,
moreover, animal dissections offer interesting possibilities of comparison.
He submits a comprehensive plan of organization for the entire animal
kingdom, and even extends his interest to the invertebrates. He also discusses
the anatomy of plants. His special zootomical investigations, which com-
prise the fourth section of his work, actually consist of a miscellany of notes
on the anatomy of a number of different animal forms; he never records the
results of a radical anatomical study of any particular animal. The chapter
entitled "Tetrapodographia' recounts scattered observations on the anatomy
of domestic animals in particular, but also of the fox, the hare, the mole,
the tortoise, and the hedgehog. The "Ornithograpbia" contains similar in-
formation on birds and a special comparative study of their feet; details
(mostly external) are given of insects and spiders, and thif zootomical hand-
book closes with a chapter on fishes, of which the ink-fish are dealt with in
greater detail. The last section of the work consists of an account of the
technique of the subject; the usual dissecting instruments are described, and
even the use of the magnifying-glass is recommended.
The title of the book, Zootomia Democritea, testifies to the tendency of
the work from beginning to end — antipathy to Aristotle, a feeling which
had been inculcated into Severino in Campanella's school. In the first
chapter he sets up the observation of nature in opposition to the theories of
Aristotle — the same system of natural observation on which Democritus
laid so much stress. Severino does not succeed, however, in creating any fresh
conception of natural phenomena in the place of the Aristotelean, and so,
like Campanella, he has to a great extent to fall back upon the mediaeval
schoolmen, whose deductive method of argument and proof he employs in
his zootomical studies.
The death-blow to Aristotle's biological theories was destined to come
from quite a different quarter; curiously enough, from a man who had the
greatest respect for his teaching, but who at the same time established cer-
tain facts which rendered it impossible for him to follow it.
CHAPTER XIV
THE DISCOVERY OF THE CIRCULATION OF THE BLOOD
I. Harvey's Predecessors
Galen s system of blood-movement
BIOLOGICAL RESEARCH Under the Renaissance, as the above narrative
shows, considerably widened the knowledge of animate nature. The
progress achieved was particularly great in the anatomical sphere-
Vesalius and his school contributed not only to human, but also to animal
anatomy a wealth of new facts which put the knowledge of classical antiquity
completely in the shade. But as regards their general conception of nature
these research-workers remained entirely on the ground that had been broken
by Aristotle and Galen. Now, however, these newly-won facts could not be
reconciled to the old system; the same thing had happened to Copernicus
and Galileo in regard to astronomy. A definite break away from the ancient
ideas of life was inevitable. In one field in particular was the influence of the
ancient system fated - regarding the idea of the movement of the blood in
the body and its importance to life. Hippocrates, Aristotle, and Galen had
all held the same views on the heart and the vessels of the body in so far as
they took the most important qualities of the blood to be the "vital spirits"
which it was thought to contain; and in face of the speculations on these
spirits the study of the movements of the blood in the veins was sadly
neglected. Galen, who among the biologists of antiquity had the richest
experimental material at his disposal, had worked up into a systematic
whole all the knowledge of the vascular system which classical antiquity
had accumulated. He had, as will be remembered, succeeded in destroying the
old illusion that the arteries and the left ventricle of the heart contained air;
he found in them a kind of blood which he believed to have acquired its'
light-red colour from the pneuma, the half-mysterious life-spirit, which it
contained. The pneuma was conveyed to the blood in the arteries from the
air, which was introduced by inhalation into the lungs and thence to the
left ventricle of the heart. The non-pneuma-conveying blood — the venous
blood — had its centre in the liver, where it was formed out of food from
the digestive canal. From the liver the blood was conveyed through the
veins partly out into the body, in which it was converted, by a process that
was not very clearly explained, into "flesh,;* and partly to the right heart-
io8
o
eq
Q
u
Pi
w
1-1
u
w
o
K-1
0
si
O
W
o
>
M
RENAISSANCE 109
chamber, from which "soot" was given off through the pulmonary arteries;
the wall between the right and the left ventricles was full of fine pores,
through which the blood oozed from the right to the left side, to be ' 'cleansed"
by the action of the pneuma. Galen had but vague ideas as to the movement
of the blood in the vessels; in the veins, at any rate, the blood moved, accord-
ing to his notion, alternately in both directions. Such was Galen's theory
of the blood-vessels and their contents, and in this form it was still accepted
by the great anatomists of the sixteenth century. All its vagueness and many
contradictions would undoubtedly have been realized long before had not the
blood-vessel system of old been considered the very centre of life itself;
the mysterious pneuma was only one side of this blood's specific life-content;
the different kinds of soul that man was believed to possess — the "vegeta-
tive," with the liver as its organ, and the "animal" in the heart — were
also intimately connected with the blood and through it affected the entire
body.^ Speculation about these components in the organism certainly did not
make the conception of the vascular system any clearer; moreover, it en-
tailed the risk that any critical discussion of these organs might be inter-
preted as an attempt to call into question the immortal soul of man, which
would inevitably have involved the scientific student in trouble with the
theologians and the Inquisition. Typical in this respect is Vesalius's attitude
regarding the pores in the dividing wall between the right and the left
heart-chambers; he could not find any trace of them, but cautiously adds
that all the same the blood might perhaps be able to ooze through the wall
itself. His pupils adopted the same cautious attitude on this point, particu-
larly Fabrizio, the discoverer of the venous valves.
To attack the traditional theory of the blood-vessels was thus a task
that required courage. The man who was the first to grapple with an at-
tempt to reform one detail of the old theory was in fact well qualified in that
respect, a man whose whole life had been spent in a struggle against time-
honoured ideas and who was at last to die for his principles. This was the
well-known religious enthusiast and martyr, Michael Servetus.
Miguel Servet y Reves, which was his real name, was born at Vil-
lanueva in north Spain, of noble parents. The date of his birth is not known
for certain (1509 or 1511); how he spent his youth is also unknown. It was
apparently at an early age, however, that he experienced that restlessness of
spirit which throughout his life made it impossible for him to settle to any-
thing permanent or find any definite mission in life. He became one of those
passionate, revolutionary, and at the same time deeply mystical enthusiasts
who were particularly in evidence during the Renaissance. Having visited
' The theory that the soul, not only of man, but also of animals, is in the blood occurs
in the Old Testament: "Only be sure that thou eat not the blood: for the blood is the life; and
'■hou mayest not eat the life with the flesh" (Deuteronomy xii. 2.3).
no THE HISTORY OF BIOLOGY
several places in Germany and Italy, he settled down in Strassburg and there
published his first treatise, De trinltatis erroribus, in which he recorded the
results of his mystical religious speculations. He disapproved of infant
baptism and expressed a view of the Trinity which was regarded as Arian.
The book evoked a storm of bitter criticism from both Catholic and Protes-
tant theologians; Server had to flee from Strassburg and subsequently re-
appeared under a different name. In Lyons he found refuge with a physician,
who inspired him with a taste for medicine, and in order to continue his
studies he moved to Paris and there practised anatomy with Vesalius. At
the same time, characteristically enough, he lectured to the students on
astrology. His theories of the influence of the heavenly bodies upon the health
again brought him into trouble with the theologians and he had to flee from
Paris. In the city of Vienne, on the Rhone, he found employment as a phy-
sician and spent there a few peaceful and happy years. During that period he
recorded the results of his continued theological speculations in a book en-
titled Christianismi restitutio. He attempted by correspondence to win over to
his views the reformer Calvin, but was rebuffed. When, in spite of this, he
dared to publish his book anonymously in Vienne and concluded it with a
venomous attack on Calvin, the latter became furious and had the author
brought before the Inquisition in Vienne. Server was cast into prison, but
managed to escape, and this time sought refuge in Geneva, probably in order
to co-operate with the anti-Calvinistic party which was just then planning
an attack on the despotic reformer. Calvin, however, was on his guard;
Server was arrested and Calvin seized the opportunity offered by the trial of
this sectarian, so hated by the whole Christian community, to strengthen his
position. Having obtained the consent of several Protestant Church councils,
the court at Geneva condemned Server to be burnt at the stake, and the ver-
dict was carried out on the xyth October 1553, to the eternal shame of Prot-
estantism. Shortly before, the Catholic Inquisition in Vienne had caused
Server's portrait to be burnt in the absence of Server himself. Through his
death, however. Server won such renown as neither his personality nor his
writings in themselves warranted; the Catholics in particular have in latter
times honoured his memory, in order to annoy the Calvinists. Statues have
been erected to him in both Paris and Madrid.
Servets investigation of the puhnonary system
Servet's principal work,^ On the Restoration of Christianity, is, as its title im-
plies, purely theological and discusses from a mystical spiritualistic point
^ Of the original edition of Servet's Christianismi restitutio there are, as far as we know,
only three copies in existence; one in Vienna, one mutilated copy in Paris, and one defective
copy in Edinburgh. An attempt to republish the work in England in the seventeen-twenties
fell through owing to the opposition of the ecclesiastical authorities. In 1790 a new edition was
at last published in Nuremberg; even this edition is somewhat rare.
RENAISSANCE III
of view the relation of God to the world and man. Every conceivable prob-
lem of life is drawn into discussion in this connexion — jurisprudence and
statesmanship as well as astronomy, physics, and medicine. In the discussion
on the Holy Spirit he points out that this cannot be properly comprehended
without knowledge of the spirit of man, and the spirit of man in turn, if it
is to be rightly understood, requires a knowledge of the human body. In
this way Server arrives at a discussion of the structure and function of the
human body, and in particular the part played by the blood, which is so
vital in its spiritual aspect. And here he pronounces the dictum that has
given him, the religious idealist, a place in the history of biology; this was
his exposition of the course of the pulmonary circulation. In order to gain
an idea of the relation of the spiritual to the physical life we must, says
Server, realize the three vital elements in the body, which are: the blood,
with its seat in the liver and the veins; " spiritus vitalis," in the heart and
the arteries; and " spiritus animalisy' which is a ray of light and is situated
in the brain and the nerves. In all these dwells the power of God's spirit.
The vital spirit is communicated by the heart to the liver, for in the heart
dwells first of all the spirit communicated by God, as we see from the embry-
onic life, in which the heart is the first point that lives. On the other hand,
the liver provides through the blood material to the spirit, which is formed
by the union of the finest components of the blood with the inhaled air. This
union takes place in the lungs, to which the blood is conveyed from the
right heart-chamber, to be conveyed thence, purged of soot through exhala-
tion and mingled with inhaled air, back to the left heart-chamber. That the
blood does not, as is commonly imagined, pass through the heart wall is
proved not only by the latter's solid consistency, but also by the powerful
structure of the pulmonary veins, which cannot be explained simply by their
function of feeding the lungs. All this is really obvious, concludes Server,
from the observations recorded by Galen, if only one understands how to
interpret them aright.
The strange, strongly spiritualistic physiology which Server expounds
in his description of the importance of the blood, above referred to, is in
itself nothing peculiar to him; on the contrary, it recurs often in the authors
of the Renaissance and even of the seventeenth century; in Swedenborg, too,
we find a similar method of speculation. Indeed, Server reminds us of the
latter in having arrived at his theories by way of speculation rather than
through his own observations. True, he had dissected, as mentioned above,
and that too under Vesalius himself, though he makes no reference to his
experiences in that line, but tries to give a correct interpretation of Galen.
What in these circumstances is surprising is that he gives such a clear idea
of the pulmonary circulation — all the more so as his view of the blood-
vessel system is otherwise purely Galenian, with the liver as the principal
111. THE HISTORY OF BIOLOGY
organ for the blood and the veins emanating therefrom. Nor indeed was it
his object to gain any knowledge of the structure of the human body for any
biological or medical purpose; just as his method was speculative, so his
purpose was exclusively mystical-theological and he thus was content with
the old tradition as it stood, except in that one point in which it was not
consistent with his metaphysical construction of thought. However, he is
undoubtedly the first to expound a theory of the pulmonary circulation
agreeing with that confirmed by the research of later times.
It is hardly to be supposed that a treatise which was prohibited by con-
temporary and later governments, and to the best of their powers suppressed,
would succeed in exercising any great influence on the progress of science;
it was, in fact, to be more than a century and a half before anyone drew atten-
tion to Servet's contribution to the discussion of the circulatory system.
Nevertheless, it would appear that Servet's ideas did after all have some
influence on his contemporaries, since during the latter half of the sixteenth
century one comes across in many authors statements, or at any rate hints, as
to the blood-circulation between the right and left ventricles through the
lungs. One or two of these writers, who had some influence on the final solu-
tion of the problem of the movement of the blood, should be mentioned here.
Realdo Columbus, Vesalius's pupil and immediate successor in the chair
of anatomy in Padua, to whom we have referred above, may claim to be
named among the forerunners in this field, as he is the only author cited by
Harvey, the great pioneer of research work on the blood. Columbus in his
work on anatomy devotes a chapter to the vascular system. Here he presents
the traditional theory of the liver as the centre of the venous system and the
true blood-forming organ, from which the blood is conveyed to the diff^erent
parts of the body. The arterial system originates in the heart. Its right and
left ventricles are separated by an intermediate wall, which, contrary to the
common assumption, is impenetrable; from the right side the blood is con-
veyed to the lungs, where it is mixed with air and, thus diluted, is conducted
back to the right side of the heart. This, he adds, no one has hitherto ob-
served or described, but it is none the less true and can be verified on experi-
mental subjects, whether alive or dead. Columbus's work was published in
1559 — that is, six years after Servet's. There have been lively discussions
whether both arrived at the same conclusion independently, and, if not, of
which borrowed from the other. The question can of course never be defi-
nitely decided, but it is probable that Servet, who indisputably has the prior
claim on the point, in some way influenced Columbus; the latter presumably
read the dangerous heretical treatise, which he dared not quote even if he
had desired to do so. There is no doubt, however, that Columbus, in a far
greater degree than Servet, confirmed his statement by observation and
experiment.
RENAISSANCE I13
There was another whose opinions on the question of the circulation
of the blood attracted far greater attention than the above-mentioned con-
tribution to the subject. This was the Italian botanist and physician Ce-
sALPiNO, who is still to this day extolled by his countrymen as the true
discoverer of the circulation of the blood. Andrea Cesalpino was born at
Arezzo in Tuscany in 15 19. He studied philosophy and medicine at Pisa,
the latter under Columbus, who was called to that city from Padua. At the
age of thirty he became a doctor of medicine and shortly afterwards professor
of pharmacology at Pisa. In this capacity he devoted special attention to
the study of botany and is reputed one of the pioneers of that science. His
contributions in this field will be dealt with in another connexion. In his
old age he was summoned to Rome, where he was appointed body-physician
to the Pope, and where he died in 1605.
Cesalpino was a man of manifold interest; besides botany and pharma-
cology he studied anatomy, mineralogy, and metallurgy, but he was above
all a natural philosopher in the true Aristotelean spirit. His theoretical spec-
ulations he published in a work with the characteristic title of Peripatetic
Problems. In this book he endeavours to find a general explanation of nature
along Aristotelean lines; in the purely philosophical aspect of his conclu-
sions he goes beyond his master by deriving both form and matter from a
single supreme principle, but as a physicist he takes his stand on the old
ground, with celestial spheres and circular planetary orbits, heaviness and
lightness as a quality of bodies — everything in fact which Galileo was
intent on demolishing.^ Even his biology, the subject of the fifth book, en-
tirely follows the lines of Aristotle. It opens with the purely mediasval scho-
lastic thesis that if the life in a being is one and indivisible, the body must
also be one and its centre one, whence life emanates to the rest of animate
things. Plants and lower animals, which are able to live even when cut up
into bits, require no such centre point, but in sanguineous animals the heart
without doubt constitutes this centre point — the heart, which is the first
to begin to live and the last to die, and which is situated in the centre of
the body. Thereupon Cesalpino endeavours, in a polemic against Galen inter-
larded with quotations from Aristotle, to prove that the veins originate in
the heart and not in the liver, and that the nerves likewise originate in the
heart and not in the brain, the latter point being proved, i7Jter alia, by the
fact that happiness and grief are felt first in the heart, while the function
of the brain is to cool the blood, like the receptacle in a distilling appara-
tus. By thus swearing to the truth of his master's word, both good and
evil, Cesalpino at any rate makes this point in regard to the circulation of
^ Curiously enough, even Cesalpino, in spite of his loyal Aristoteleanism, fell into the
hands of the Inquisition, but he saved himself by his dialectical cleverness, and perhaps also
owing to his being in the papal service.
114 THE HISTORY OF BIOLOGY
the blood, that the heart is actually the centre of the vascular system. And
in regard to the relation of the lungs to the heart, he maintains with his
teacher Columbus that the blood passes through the lungs from the right to
the left side of the heart — a process which he for the first time calls cir-
culation. But his servility to the authority of Aristotle prevents him from
taking advantage of either his precursors' or his own progress in this field
of research. He dares not abandon the theory of the pores in the heart wall,
but, on the contrary, admits that some of the blood goes that way; he ob-
serves that when a vein is tied, it fills below and not above the ligature,
but he does not venture to draw the conclusion that the blood-stream in
the veins always leads to the heart — this he believes takes place during
sleep, but not in a waking condition — and so the existence of the "vital
spirit" in the blood naturally takes the first place in his investigations. His
ponderous and involved presentation of his case — vague, too, in compari-
son with Servet's brief and explicit style — has enabled his admirers to
interpret his statements as it suits their purposes, but just as none of his
contemporaries saw in him one who had revolutionized knowledge of the
vascular system — a fact which he himself, Catholic and papal favourite as
he was, would hardly have dared to admit — so there must in truth be a
partisan and chauvinistic spirit in those of posterity who would ascribe to
him the honour of an idea which he himself neither clearly expressed nor
ever definitely claimed.
X. Harvey
Besides those of whom we have given account above, there were during
the Renaissance, as has been said, quite a large number of anatomical writers
who made a study of the construction and function of the vascular system,
in vain attempts to bring order out of the chaos to which the inaccurate
conception of the ancient biologists had reduced the problem. The necessity
of a solution was generally acknowledged; several had been on the right
road, but had stopped prematurely. Then William Harvey took the decisive
step and solved the hard problem in one stride.
William Harvey was born at Folkestone, on the south coast of England,
in the year 1578, of respected and well-to-do parents, who gave their children
a sound education. Having taken a philosophical degree at Cambridge,
Harvey made a number of journeys and eventually came to Padua, where
at that time Fabrizio had begun to attract pupils from far and near. Harvey
joined them, and after four years of study took the degree of doctor of
medicine. Returning to England, he settled down in London and started a
medical practice. He practised in hospitals, was elected a member of the
RENAISSANCE II5
London College of Physicians, and there gained such a reputation that he
was commissioned to give lectures to his colleagues. Eventually he was ap-
pointed court physician to King James I and later to King Charles I. After-
wards he spent many years in peaceful and uninterrupted research work and
in the duties of his medical practice in London, but then the great Civil
War broke out and Harvey accompanied his king in his flight from London,
while his house was plundered and his collections destroyed. He was then
made a professor at Oxford, which was the headquarters of the King; when
this city was also captured by the Parliamentary army after Charles's final de-
feat, Harvey, then sixty-eight years old, had to retire into private life. Fortu-
nately he possessed private means and was also supported by his brother,
a wealthy London merchant, so that his old age was free from care, and at
the same time he retained the deep respect of his countrymen and colleagues.
A stroke brought his life to a sudden and peaceful close in the year 1657.
He left his fortune by will to the College, whose leading personality he had
been during his lifetime, and ever since his death the College has continued
to celebrate his memory, an annual festival being held in London in his
honour. A fine monument has been erected over his grave.
Harvey's ivork on the circulation
The work in which Harvey expounded his new idea of the circulation of
the blood was published in i6i8 at Frankfurt am Main in the form of a
quarto volume containing seventy-two pages. Harvey, however, had spent
his whole time, ever since, as a youth, he received his first lesson in anatomy
in Fabrizio's school, in working out the ideas which were recorded in this
modest volume. There are still extant the lecture notes dating from 1616,
in which are expressed some of the thoughts which twelve years later as-
sumed their final form, and it has thus been possible to check the careful
research, the mature consideration, on which the work is based and which
shows itself in the masterly style, at the same time concise and explicit, in
which not a word seems superfluous. After giving an account of the old tradi-
tional theories on the subject, in which he sharply brings out their defects,
Harvey presents his own observations on the movement of the heart. Ac-
cording to the old theory the walls of the heart were not muscular and the
dilatation of the heart was its most important function; by this means the
blood was conveyed from the veins into the heart. By careful experiments,
of which he gives an account, Harvey found that the heart is muscular and
that, on the contrary, its regular contraction is its most important move-
ment, which drives the blood forward — that is, out into the blood-vessels
— just as it is likewise during this movement that the heart beats against
the thorax. In this movement not only the ventricles of the heart take part,
but also its vestibule, the significance of which Harvey rightly emphasizes
for the first time. He then gives an account of the course of the blood from
Il6 THE HISTORY OF BIOLOGY
the right to the left side of the heart through the lungs, ar.d in this he ac-
knowledges the services of Columbus in his explanation of this phenomenon.
With regard to the part played by the lungs and the air in this circulation
he has not much to add to the hypotheses of his predecessors. After having
thus described the small circulation Harvey proceeds to a presentation of
the blood's movement in the body itself and it is here that he brings out
his most daring originality. According to the old theory, food was converted
in the liver into blood, which was driven through the veins partly to the
heart, in order to receive the " spiritus vitalis," and partly into the body. To
this theory Harvey opposes a mathematical calculation; if the human heart
contains two ounces of blood and gives sixty-five beats to the minute, then
it drives in less than one minute ten pounds of blood out into the body.
Such a quantity of blood cannot incessantly arise from the food consumed,
but it must be assumed that the same quantity of blood incessantly circu-
lates in the body; it is driven out through the arteries and returns through
the veins. Harvey then collects a quantity of evidence in proof of this con-
clusion from the relation of the arteries and the veins in the body. He in-
vestigates the arterial pulse both in normal individuals and in those having
calcinated veins; he opens a live serpent and ties up first the vetia cava and
then the aorta; while the vein is emptied between the heart and the ligature
and swells up on the other side, the contrary is true of the aorta. He studies
the venous valves in a man's arm, which were discovered by Fabrizio, and
shows how they swell below a ligature; he severs a vein and an artery par-
allel to it and shows that the blood flows from the different ends of the
wound. On these and several other grounds, deduced from the study of every
possible animal form, he draws the conclusion that the arteries convey the
blood from the heart out into the body; there it is transmitted into the rami-
fications of the veins and flows from these into the principal vein and thence
back to the heart. The arterial blood, he considers, provides nourishment for
the body, while that of the veins is impure. How the transition between the
arterial and venous system takes place he could not explain; the capillary
system he was unable to distinguish, not having access to a microscope, and
he therefore assumed that some kind of ramified hollows formed the con-
necting link between the two. Another weak point in his theory was that
he could never find a satisfactory explanation of how the components of
the food are converted into blood, but he had to be content with the old
hypothesis that the liver was the medium in this process. He lived to see
the discovery by others of the lymphatic and thoracic ducts, but then he
was no longer capable of realizing how well these experiences complemented
his own discoveries; he desired to know nothing about them and on this
point adhered to the old theory.
If we compare Harvey's account of the circulation of the blood with
RENAISSANCE II7
the old vascular theory, we find fundamental differences in the two concep-
tions, both anatomically and physiologically. According to the old theory
the heart was not a muscular organ; it dilated purely passively and allowed
the blood to enter in order to be provided with " vital spirit," this being the
primary life-function of the heart, if it were not also, as Aristotle and
his followers until Cesalpino held, the centre of intelligence. Again, the
blood moved of itself owing to the specifically living qualities which the
" spirifus" lent it. Harvey, on the other hand, proves that the movement of
the blood is due to the purely mechanical function of the heart: the heart's
muscular contraction propels the blood out into the blood-vessels, through
the arteries out into the body, thence back to the heart through the veins,
and so farther through the lungs. In this contrast lies, one may say, the great
ditference between the ancient and the modern biological conception. Even
Harvey's way of producing his proofs is purely modern; while Servet still
refers back to philosophical speculations and the interpretation of classical
authors, Harvey propounds a purely mathematical calculus on the volume
of the heart and vascular system and continues to prove his thesis by means
of observations and experiments on a number of both higher and lower ani-
mal forms. He thus fulfils in the sphere of biology the requirement which
his contemporary Bacon laid down as a principle of science: to explain na-
ture by experience based upon observations and experiment. And even Gali-
leo's fundamental principle governing natural research — to measure what
can be measured and to make measurable what cannot be measured — is
applied to living nature by Harvey for the first time. Galileo also thought
that science can only explain how the forces of nature operate; what their
innate essential quality is will never be known under any circumstance. In
his explanation of the circulation of the blood Harvey does indeed fulfil the
first half of this principle; on the other hand, by adhering to the ancient
belief in the vital spirits in the blood he remains in his theoretical concep-
tions entirely on ancient ground.
On the generation of animals
This conservatism of Harvey's displays itself conspicuously in a work which
he published in his old age: Exercitationes de generatione animalium (165 1).
Like his work on the circulation of the blood, this book is the fruit of many
years' labour, but in contrast to the former it is somewhat lengthy and far
less perfect in form. In this Harvey gives a comparative account of the em-
bryonic development in higher and lower animal forms. He is able here to
quote as his precursor his old teacher Fabrizio, and he does so with the
utmost piety. But above all he proves himself to be a follower of Aristotle,
whose conception of the true essence of life he has made entirely his own.
Along Aristotelean lines he endeavours to find a formal unity in the mani-
fold aspects of phenomena, as displayed in the evolution of the embryo, and
Il8 THE HISTORY OF BIOLOGY
he believes that he has discovered such unity in the egg, out of which all
living creatures are evolved. His dictum: "All animals, even those that pro-
duce their young alive, including man himself, are evolved out of the egg"
is well known. He was naturally not able to observe the eggs of mammals —
such a study requires a microscope, which he did not possess — but he pre-
supposes their existence on theoretical grounds, a conclusion which was
confirmed long afterwards. Nevertheless, he is loath to abandon the idea
of primal generation, though he limits this principle to the lowest animals.
Out of the egg the higher animals are evolved by epigenesis, in that the
organs are successively formed out of the indifferent matter in the egg, which
thus constitutes, in harmony with Aristotle's theory, the potentiality out of
which the individual is realized. The lower animals, on the other hand, are
evolved by metamorphosis, a direct reconstruction of complete rudiments,
as is proved especially by the evolution of the pupa of insects; Harvey, in
fact, shares Aristotle's belief that the pupa is the insect's egg. On the sub-
ject of reproduction his ideas are entirely mediaeval; the influence of the
sperm on the development of the embryo he believes to be due to the vital
force it contains, and this he compares with the secret force exerted by the
heavenly bodies upon all life on the earth. That this last work of Harvey's
should also contain a mass of remarkable detailed observations is not sur-
prising; he describes with unprecedented care the ovary of the hen and its
development, the nourishing of the chicken in the egg, and its growth from
the very earliest stages; and of even greater interest are the comparisons he
makes between the embryonic stages in different animals — mammals, birds,
and lower types.
Harvey is without doubt one of the most remarkable figures in the his-
tory of human culture. His work is the most revolutionary that the develop-
ment of biology has to show, for it undermines the foundations of the ancient
conception of life and its manifestations, and nevertheless he himself retains
this very conception as long as he lives. He thus brings to a close the great
epoch in the history of biology which is governed by the ancient conception
of nature and he initiates the modern development in the sphere of biology,
just as Galileo does in that of physics. How an entirely new science of life
has developed on the foundations laid by Harvey will be shown in the next
section of this work.
PART TWO
BIOLOGY IN THE SEVENTEENTH
AND EIGHTEENTH CENTURIES
CHAPTER I
THE ORIGIN OF THE MODERN IDEA OF NATURE IN THE
SEVENTEENTH AND EIGHTEENTH CENTURIES
CLASSICAL ANTIQUITY gave fise to two explanations of natural phenom-
ena, each splendid in its own way: that of Democritus and that of
Aristotle. As will be remembered, Democritus attempted to explain
all phenomena in existence, both physical and psychical, by the assumption
that things were composed of a mass of particles, varying in size, shape, and
movement, whose mutual interrelation caused all that is and all that
happens, all, in fact, that is observable or conceivable. The weakness of this
theory lay in the fact that it gave no explanation of the obedience to law
which experience has proved beyond any doubt to exist in all that happens
in nature. It was therefore supplanted by Aristotle's cosmic explanation,
which maintained just this universal obedience to law, but based it upon the
assumption of a divine intelligence which governs and gives form to what is
in itself formless matter, controlling the latter in various degrees — less in
inanimate nature, more in the animate, and most in the celestial spheres
which hold sway over the imperfect earth. In animate nature this force ap-
pears as soul, vital spirit, which creates higher forms of existence the more it
overcomes matter. This cosmic theory, which, owing to its logically consis-
tent formulation, is unique in its greatness, has been characterized as dynamic
and vitalistic in contrast to materialistic atomism. It has with greater reason
been called aesthetic, since Aristotle really looked upon natural phenomena
from the point of view of an artist who gives form to matter; it has even been
called teleological, because according to it everything in existence has a pur-
pose which is determined by the governing intelligence. In this latter charac-
teristic we really find that quality in the Aristotelean thought-system which
has proved most fateful both for that system and for man's conception of
life in general. The divine intelligence which Aristotle invented in order to
make possible the assumption of law-bound existence on purely speculative
grounds became a welcome ally to the pious aims of late antiquity and still
more so to the mediaeval Church. One found indications of similarity be-
tween it and the "divine power" of the old myths of creation, and thus
received an idea of the course of the world, apparently scientific, but actually
based upon legends from the childhood of man.
I2.I
I-LZ THE HISTORY OF BIOLOGY
Aristoteleanism in alliance ivith the Church
It was just on account of its semi-scientific nature that it was extremely
difficult to controvert this idea with reasons and proofs, seeing that it was
at the same time cherished by the authority of the Church and protected by
the latter's powerful resources, both spiritual and temporal. In the first part
of this work it has been shown how the champions of natural science during
the Renaissance took up the cudgels against Aristoteleanism, which was
upheld not only by the authority of the Church, but also by the boundless
respect that that age entertained for antiquity; how Cusanus and Bruno
asserted the infinity of cosmic space in contrast to Aristotle's spherical uni-
verse; how Francis Bacon ruthlessly exposed the abstract structural system
of the ancient philosophy and urged that research should be based on obser-
vations of nature itself; how Galileo, by means of observational material
and mathematically conclusive proof, demolished the theory of the immu-
table regularity of the celestial regions and at the same time proved the
purely mechanical obedience to law of the phenomena of motion here on
earth; how Harvey through his discovery of the circulation of the blood
proved a purely mechanical action in the life-process which the old theory
considered to be the centre of animate life. But however many defects could
be proved against the old system in detail, it nevertheless still remained
unaff"ected, owing to its consistently carried-out construction; it required an
entirely new system of thought in place of the old before the latter could
definitely break down. Throughout the seventeenth century keen-minded
thinkers set about creating such a system, and the strength that underlay
Aristotle's cosmic idea, its unassailable consistency and perfect lucidity, had
never been demonstrated so clearly as now, when, already condemned to
fall, it made a stand against the assaults which finally shattered it. A survey
of this struggle between Aristoteleanism and the new systems of thought is
so much the more necessary as an introduction to the history of modern biol-
ogy as it was actually during this struggle that not only the natural science
of our own time, but the whole idea of life as conceived by present-day
humanity in general came into being. Nevertheless the modern conception
of nature by no means rests solely upon the purely mechanical foundations
laid by Galileo. It is self-evident that in it there are very considerable ele-
ments of vanquished Aristoteleanism. But besides him there appeared also
in opposition to the latter theory philosophers who from neo-Platonism and
other similar systems of ideas adopted a purely mystical view of nature.
This too has possessed its attractive sides for the human mind, especially the
advantage of making possible a uniform conception of both the material
and the ideal aspects of existence; during the Renaissance in particular it won
many adherents — Bruno is the most brilliant example — and it has con-
sequently left a strong impression upon modern natural science. In the fol-
lowing chapter an attempt will be made to illustrate how these elements in
the conception of nature in our own time arose.
CHAPTER II
THE MECHANICAL N A T U R E - S Y ST E M S
The period of the great systems
THE SEVENTEENTH CENTURY has been Called the period of great systems
of thought, during which all the knowledge that the Renaissance
brought to light was summarized and classified. Order and system
were, in fact, what this epoch strove to create in every sphere of life; in
government the power was concentrated in the hands of despotic princes
who by a rigorous exercise of power overcame all opposition on the part
of their subjects and created ordered forms of administration instead of the
universal unrest of the Middle Ages and the Renaissance; in the religious
sphere the different denominations combined into stable churches which
prohibited any divergence from the strictly formulated dogmas which they
set up. Such a period was bound to be devoted to strictly delimited systems
even in the scientific field, and indeed many such systems of different trends
of thought, but all definitely formulated, more dogmatic than critical, based
upon speculation rather than upon observation, saw the light of day during
this epoch. Some of these which highly influenced the development of
biological science deserve further mention.
The pioneer of modern philosophy
Rene Descartes (in his Latin writings he calls himself Cartesius) is com-
monly regarded as the pioneer amongst the systematic philosophers of the
seventeenth century. Born in 1596 of wealthy parents, he was able to devote
his whole life to research. His home was in Brittany, and he was brought
up by Jesuits; he spent some years in Paris and was for a time an engineer
officer in the service of foreign powers. In order that he might devote himself
to his science undisturbed by the Catholic Church, he eventually settled
down in Holland, where his most important work was done. He made a
journey to Sweden, and died in Stockholm in 1650.
Descartes, like Pythagoras and Plato, was a mathematician, and like
them, too, addicted to abstract speculations. His ambition was to place sci-
ence on firm ground, valid for all possible phenomena, and excluding all
accidental circumstances. Among these latter he counted, above all, mental
impressions, and in order to exclude them he resolved to doubt everything
in existence. But the very fact of doubt proved that he thought, and thinking
gave him proof that he existed; " Je pense, done je suis" was his oft-quoted
113
114 THE HISTORY OF BIOLOGY
Starting-point. On tiiis foundation he then builds up his entire conception
of existence on the principle that the composite should be explained from
its simple components. First he constructs out of the thought of man an
idea of God, since man's finite and imperfect personality presupposes an in-
finite and perfect origin; once we believe this, we must also be right in rely-
ing upon our mental perceptions, for God cannot have given them to us
without cause. Through the senses we are convinced that matter exists. The
simplest and therefore the most essential qualities of matter are extension,
divisibility, and mobility. On the other hand, form, which Aristotle, it will
be remembered, made his main principle, is of momentary and therefore of
secondary significance. Descartes also rejects the atomic theory, for it is in-
consistent with the principle of divisibility; nor does space exist, for every-
thing that exists must have extension. On these principles — extension,
divisibility, and mobility — Descartes bases the whole of his theory of mat-
ter, both inanimate and animate, and he entirely rejects the theory of final
causes, for it would be presumptuous to ascribe any limited purposes to
unfathomable and infinite divinity. The only rational explanation of the
universe is to regard the whole as a machine. Through vortical movements
within the parts of matter the latter have accumulated and become heavenly
bodies; and movement is all that takes place in nature.
Life-phenomena -purely mechanical
On this same principle he seeks to explain the phenomena of life — that is,
the corporeal. These, in his view, occur purely mechanically, without the
intervention of any of the spiritual forces which the Aristoteleans assumed,
whether animal or vegetative. Confirmation of this idea of the living body
as a mechanism Descartes found in Harvey's discovery of the circulation of
the blood, which he enthusiastically upheld and to the acceptance of which
he powerfully contributed. This fact in itself would be sufficient to ensure
him a place in the history of biology. And in drawing conclusions from
Harvey's observations which the latter, faithful Aristotelean as he was,
could himself never have perceived, he formed a theory of the human body
as a mechanism which may be regarded as the foundation of modern physi-
ology. In particular, he sought to explain mechanically the function of the
nervous system. He believed that from the brain the so-called animal spirits
are conveyed through the nerves to the muscles, which are thereby set in
motion through the impulse given them from the brain. It is not at all neces-
sary that these impulses should be conscious; they may take place in the com-
plete absence of thought " just as in a machine." Thus mental impressions
can immediately call forth movements through the nerve currents' being
"thrown back" — that is, reflected. He has thus recognized, described, and
from his own point of view explained the phenomenon of reflex motion.
In regard to animals he believes that all their manifestations of life are the
SEVENTEENTH AND EIGHTEENTH CENTURIES 1x5
result of such reflexes; it is not possible, he thinks, to ascribe to them the
possession of a soul. That man has a soul, on the other hand, Descartes con-
sidered to be proved by the fact that man has consciousness, and this soul
he regards as a substance, the existence of which, however, is entirely inde-
pendent of the body. Only at one point, he considers, is there any co-opera-
tion between soul and body — namely, in the glandula pinealis; there the
currents from the nervous system react upon the soul and impart to it a share
in mental impressions; there, on the other hand, the soul substance makes
impressions upon the nervous system, which give rise to conscious actions.
Thus Descartes has created a purely mechanical cosmic theory in which
everything that happens takes place out of mathematical necessity; in which
neither the accidental movement of the atoms nor the direct intervention of
God is needed to keep the course of events on the move. It is manifest that
the great discoveries made during the Renaissance have conditioned his
theory. But he himself would not acknowledge any precursors — his ac-
knowledgment of Harvey constitutes the one exception — he was, in fact,
a very cautious man, and Galileo's fate had made a deep impression on him.
He anxiously avoided offending the Church, for which he always showed a
deep respect; his manner of escaping from controversy was more adroit than
courageous, as when he gives an assurance that his theory of the creation is
merely a game of thought; it might be conceivable that the universe arose
as his theory declares, but one knows all the same that the Church main-
tains the true theory of creation. This, together with his constant assertion
of the immortality of the soul of man, was, however, the reason why so
many eminent ecclesiastical personages dared to embrace his theory, and thus
the mechanical explanation of the cosmos wormed its way, one might say,
into the consciousness, thrusting out Aristoteleanism. And naturally the new
explanation of the cosmos — Cartesianism, as it was called — had great ad-
vantages over the old — above all, in that it rendered possible the appli-
cation of the newly-achieved results of research in the fields of physics,
astronomy, and biology. But even the new theory had its weak points. In
particular, the relation of the consciousness to material phenomena, or, in
other words, the soul's relation to the body, was a problem which worried
Descartes and which, as we have seen above, he finally solved, though not
very successfully. For Aristotle this problem did not exist; he had in fact
made the soul equivalent to form and thus evaded the point. For the rest,
it seems that the individual life did not concern him very much, any more
than it did the other philosophers of early antiquity. Late antiquity, and in
a still greater degree Christianity, had, on the other hand, devoted earnest
attention to this problem, and now that material phenomena were givr.n a
purely mechanical explanation, it became extremely acute and was for a long
time to be the main point of discussion in the philosophical agenda. This
1X6 THE HISTORY OF BIOLOGY
question has naturally been of only indirect importance to biology, which
deals mostly with the material phenomena of life, but all the same it has
had its influence on many purely biological problems and therefore cannot
be entirely neglected here.
It was just in this sphere of science that Cartesianism experienced the
strongest opposition on the part of such scientists as did not accept Aris-
totle's views. In France it was Pierre Gassendi (1591-1655) who stands out
most conspicuously as the opponent of Descartes. He was born of poor par-
ents, but rose to high dignities in the Catholic Church. As a philosopher he
sought to revive the ancient atomic theory and wrote a defence of Epicurus,
who in the Middle Ages was the object of universal execration. Against
Descartes he argued that his conclusion that thought was a proof of exist-
ence was incapable of realization. For the rest, Gassendi was a great admirer
of Galileo for his discoveries in the realm of physics, which he partly im-
proved upon; being a priest, however, he was forced to deny the Coperni-
can cosmic system. He conceived warmth to be the soul in existence. The
relation between matter and human consciousness he tried to explain in the
same way as Lucretius, but he admitted that there were insoluble difficulties
in the way; besides, as a priest he had of course to maintain the existence
of an immortal soul.
Another thinker who held a markedly mechanical view of existence was
the Englishman Thomas Hobbes. He had studied in Oxford and travelled a
great deal in Europe, and afterwards spent the greater part of his long life
as a private scholar. He died in 1679 ^^ ^^^ ^E>^ ^^ ninety-one. He regarded
all that happens as motion; mental impressions were in his view motions in
the nervous system, which arose as a reaction to motions in the external
world. Hobbes speculated most, however, upon problems of ethics and states-
manship; he had no interest in biology.
The same may be said of another philosopher, who nevertheless, curi-
ously enough, came to play a not unimportant part in the history of biology
— Baruch Spinoza. Born of Jewish parents at Amsterdam in 1631, he was
brought up to become a rabbi, but as he failed to follow the teachings of
the synagogue, he was excommunicated and afterwards lived in the closest
retirement, making a livelihood by polishing eye-glasses, until the time of
his death, in 1677. Only a few liberal-minded people dared during his life-
time to acknowledge acquaintance with the outcast, and although in some
respects he was highly admired, it was not until later that his writings won
any general acceptance. He himself, thanks undoubtedly to his mild and un-
assuming temperament and his retired life, escaped falling a victim to the
religious fanaticism of the age, for even in Holland, which was a compara-
tively liberal-minded country, tolerance towards heterodox persons was a
rare thing in those days.
SEVENTEENTH AND EIGHTEENTH CENTURIES II7
Spinoza's system of thought is one of the most magnificent and consum-
mate in human history, one of the most ingenious attempts to reconcile the
opposition between consciousness and matter which Cartesianism brought
out. In it he is governed by a feeling inspired by the Jewish faith of his
childhood — a religious awe of the infinite, eternal, immutable, "that
which is in itself and is comprehended out of itself." This he names sub-
stance: that into which all things that exist enter as parts. This immuta-
ble substance has an infinite number of forms in which it appears, of which
we human beings can distinguish only two: the material extension, and the
spiritual consciousness. These cannot in any way be explained out of one
another, but, on the other hand, both revert to the substance out of which
they arose, and man can therefore conclude from the laws that govern the
one that they also govern the other; the laws of human reason have abso-
lute force in nature as well. From the immutability of the substance it
follows that the development that seems to take place is only apparent;
everything, after a brief individual existence, reverts to the substance, like
a wave that sinks back into the sea, giving place to new individuals of
equally ephemeral existence. To acquire knowledge of the substance is the
highest aim of man; it cannot, however, be attained by way of thought, but
only by direct introspection. Spinoza thus ends in mysticism — that, too,
probably induced by his Hebraic-oriental origin. It is strange that, in spite
of this and of his utter denial of any kind of development, his system has
been deeply admired by the very students of nature of more recent times who
have made development the principal aim of their researches. Goethe was
strongly influenced by it, and in more recent times Haeckel and his monist
disciples have given it enthusiastic support, in reality perhaps more on ac-
count of the religious persecution suffered by Spinoza than on account of
the subject-matter of his extremely involved writings.
In most respects his somewhat younger contemporary Gottfried Wil-
HELM Leibniz forms a sharp contrast to Spinoza. Leibniz was born at Leipzig
in 1646, the son of a professor. He was a veritable infant prodigy; as a boy
he had read practically the whole of the classics and at the age of seventeen
he delivered his doctor's dissertation. Mathematics and jurisprudence were
subjects of particular interest to him; he became one of the pioneers of the
former science, while the latter provided him with an income as a govern-
ment official and diplomatic representative at the courts of several German
minor princes. He died at Hanover in 1716. Throughout his life his energy
was remarkable and his interests incredibly many-sided. In the course of his
travels in most countries with any standard of culture, he had made the ac-
quaintance of the most eminent men of his time, and in questions of culture
his advice was sought from all sides. Peter the Great of Russia, as well as
the most learned men in western Europe, corresponded with him. And as his
1X8 THE HISTORY OF BIOLOGY
temperament was as pacific as his interests were universal, he endeavoured
everywhere to reconcile and to unite. At one time he speculated upon a
universal science in which all human knowledge was to be represented by
short symbols; on other occasions he worked for the union of the different
Christian Churches. The same efforts to reconcile opposed views likewise
govern his natural philosophy. Thus, he seeks to show that the Church's
doctrine of the omnipotence of God, and natural science's mechanical ex-
planation of the universe are by no means mutually exclusive, but are capable
of being harmonized.
Leibniz.' s raonad theory
As a natural philosopher Leibniz took the atomic theory as his starting-point.
As he found it impossible, however, to derive consciousness and the mani-
festations of the soul in general from the movements of atoms, he sought a
way out of the difficulty by assuming a universe composed of units of an ideal,
not of a material, character. The idea for this theory he found through using
the microscope, which had then just been invented; by this means it is
possible to see that every drop of water swarms with life and that life exists
everywhere, even where the eye cannot see it; it was thus but a short step
to the conclusion that the smallest particles of matter are life-principles —
not dead atoms, but living "monads," as Leibniz called them. These monads
he conceives as being of infinite variety, some of a higher type, others of a
lower. The human soul is one such monad, which has consciousness; the life
of animals consists of lower monads, unconscious, but percipient; the monads
of plants live, but are not percipient; the monads of inanimate nature are in
an indifferent state, as in a dreamless sleep, the human body being composed
of these latter monads. The activity of the monads is not motion, as the atomic
theory supposed, for motion is something relative, but their ultimate quality
can only be conceived as force — conatus, as Leibniz calls it. By this means
they each, in a higher or lower degree, obtain some notion of existence. On
the other hand, they do not react upon one another, their interrelation being
governed by a harmonious cosmic order, originally created by God, who is
the supreme monad. Thus, the human body functions by force of the harmony
of existence parallel and in tune with the soul, like two clocks which go
exactly alike. The kingdoms of nature and of grace act similarly towards one
another. — All this extremely abstract speculation might at first sight appear
foreign to all that is meant by natural science. Leibniz has, however, actually
exercised great influence on natural research, partly by awakening interest
in life, both in the great multiplicity of its manifestations and in its most
minute forms, and, above all, by insisting upon the idea of force as the basis
of natural phenomena instead of movement, which even Descartes believed.
And his endeavour to reconcile the kingdoms of nature and of grace, which
may appear foreign to the ideas of natural research of our own day, would
SEVENTEENTH AND EIGHTEENTH CENTURIES 119
have seemed by no means unattractive at a time when the foremost natural
philosophers were at the same time pious Christians and subservient each
to his own Church. This even Galileo had been, as also those two of his
successors who have contributed more than any others towards giving natural
research its modern character — Boyle and Newton. These two are all the
more worthy of mention here as through their activities they have, each in
his own way, powerfully, if only indirectly, affected the development of
biology.
Robert Boyle (1617-91) is generally looked upon as the first modern
chemist, in so far as he definitely broke away from the mystical speculations
of alchemy and made the object of chemistry the breaking up of complex
substances into their simplest elements. Thus he freed experimental effort
from the fantastic and semi-magical aims and means of the Middle Ages and
created a natural-scientific method based on rational calculations. On the
other hand, he had little bent for purely speculative problems and accepted
the general cosmic viev/s of the Church without reserve.
Far more renowned and of far greater influence on the intellectual prog-
ress of man was Isaac Newton, one of the greatest pioneers of natural science
that the world has seen. Born in 1641, of a peasant family, he studied in
Cambridge, was for many years professor there, became in his old age Direc-
tor of the Royal Mint in London, and died at the age of eighty-four, honoured
and respected as few scientists have been. He was known everywhere for his
liberal-minded political views, deep religious sense, and modest, lovable
nature.
The gravitation theory
Newton's important discoveries in the sphere of mathematics and optics are
universally known. Most famous and most vital from the point of view of
cultural development is, however, his theory of gravitation. Galileo had, it
will be remembered, established the fact that the movement of bodies on our
earth takes place on fixed, mathematically calculable principles. Newton
now proves that the same laws governing the movement of bodies at the
earth's surface also govern the movement of the heavenly bodies in their
relation to one another. All the world knows the story of how in his youth,
at the sight of a falling apple, he began to ponder the question whether it was
possible to calculate the movement of the moon round the earth according to
the same law of attraction as that governing the fall of the apple. He spent
twenty years working out his idea and finally laid down the well-known
principle that bodies attract one another with a force directly proportional
to the mass and in inverse ratio to the square of the distance. The extraor-
dinary importance of this discovery was by no means immediately clear to
everyone; it was at variance with the speculations of the Cartesian philoso-
phers as well as with the doctrines of the theologians, and it was not until
130 THE HISTORY OF BIOLOGY
after the lapse of several decades that humanity realized that here was a new
foundation on which to base the conception of the universe. This was, of
course, due partly to the fact that Newton himself remained in certain re-
spects at a somewhat antiquated point of view. Like Galileo, he was quite
aware, as he himself says, that it has not been possible to discover "the cause
of the qualities of gravitation from phenomena, and I form no hypotheses,"
and he maintained that it was justifiable to conclude from the existence of
properties in those bodies which had been investigated the existence of the
same properties in all bodies. At the same time he was firmly convinced that
finality in nature presupposes a personal God as the Creator of the universe,
and even as the Maintainer of the whole, since the irregularities in the course
of the heavenly bodies must some time be adjusted through the personal
intervention of the Creator. The latter assumption, which induced Leibniz
to liken Newton's cosmic system to a clock which now and then had to be
regulated in order to go properly, testifies more clearly than any other factor
to that mixture of childlike innocence and intellectual keenness in Newton
which gives to the whole of his personality the character of old and new in
conjunction, such as is so often to be found in philosophers at the turning-
point in the scientific history we are here discussing. It was left to the eight-
eenth century entirely to shake off the traditional ideas of the structure of
the universe and in their place to create that theory of existence which has
been maintained ever since. The man who more than any other exerted a
decisive influence in this respect is usually, and rightly so, not counted a
scientist at all, yet he has in a greater degree than most aff'ected the progress
of science; that man was Voltaire.
pRANgois Marie Arouet de Voltaire is one of the best-known and most
discussed figures in cultural history — uncritically vaunted to the skies by
his admirers, violently calumniated by his enemies. In the course of his long
life (1694-1778) he exercised, more than perhaps any other has ever done, a
purely cultural influence in every sphere of life. His literary, political, and
religious activities are universally known. As a man of letters of middle-class
origin he had acquired a name in Paris for his cleverness and love of opposi-
tion, when he was suddenly ordered by the Government to leave the country
and took refuge in England. After a three years' sojourn in that country
(172.6-9) he returned full of ideas which he had assimilated there, and de-
voted the rest of his life to making them known to the world. Among these
new conceptions were Newton's discoveries in physics and astronomy. By
combining these with a number of ideas gathered from Leibniz's speculations
he produced a theory of the universe which was not only purely mechani-
cal — that the Cartesian theory had already been — but which was also
based upon mathematically calculable facts and was therefore bound to work
with indisputable authority. He worked with indefatigable enthusiasm to
SEVENTEENTH AND EIGHTEENTH CENTURIES 131
get this theory known, using his brilliant literary powers and popular gifts
to make it attractive; indeed, it was mostly thanks to him that Newton's
discovery became known to the world of culture in Europe within a space of
a few decades. Thus it came about that Voltaire to a certain degree stands in
the same relation to Newton as Haeckel does to Darwin. Voltaire also re-
minds us of Haeckel in that he made his natural-scientific theory the basis of
a comprehensive view of the world, to which he unceasingly refers in his
struggle against ecclesiastical authority, whose doctrines of the creation
and of miracles he despised and ridiculed from that standpoint. From his
time originates the custom of citing "natural laws" as proofs controverting
the Church's traditional cosmic theory. Otherwise Voltaire's notions of the
universe constitute in themselves no really radical break with the old tradi-
tion; he believed both in a personal God and in the causal finality of nature,
which to a certain extent contributed towards making the transition from
the ancient to the modern cosmic theory less of a shock to the great majority.
Furthermore, his doctrines had the rare consequence of bringing about a
radical revaluation of the whole of man's ideas of life.
With the coming of this so-called "period of enlightenment" introduced
by Voltaire we may regard the conception of nature created by antiquity
and handed down through the Middle Ages and after as definitely shattered.
It would, however, be an exaggeration to assume that Voltairianism reigned
supreme in his period. Besides the adherents of the time-honoured cult of
antiquity, of whom there were always a great number, there were to be
found during the enlightened period, in ever-increasing numbers, supporters
of the mystical-speculative tendencies which have been mentioned above.
Throughout the whole era here under discussion a not unimportant part
was played by natural-scientific mysticism, influencing even otherwise quite
critical people in the scientific world and everywhere attracting adherents,
who devoted themselves entirely to its aims and purposes. Its roots, as has
already been mentioned, lay far back in time, while its ramifications can be
traced even in the field of modern natural research, exact though it apparently
is. Its development thus deserves a chapter to itself, the beginning of which
must take us back to the days of the Renaissance.
CHAPTER III
MYSTICAL SPECULATION UPON NATURAL SCIENCE
Magic during the Renaissance
IN THE FIRST SECTION of this work wc havc pointed out how important was
the role played by mystical speculation in science during the Renais-
sance, even in the theories of its principal representatives, such as a Cusa-
nus or a Bruno. As a matter of fact, through the break-down of scholasticism
in science, the field was left open for all those wild fantasies that seem to be
common to all times and generations, although they are at times thrust out
of sight and dare not show themselves for fear of learned authority and the
derision of critics. Seldom indeed is it that mystical speculation and magical
experiments have gone so far and had such scientific pretensions as during
the Renaissance.
All this Renaissance magic was based on a great many preconceptions:
primitive superstition, cabbalistic interpretation (originating from the East)
of the books of the Bible and a number of apocryphal appendices, Arabian
experimental science and its Western development. Finally, the neo-Platonic
philosophy, striving to gain by means of direct introspection a mystical, uni-
form conception of the whole of existence — spirit and matter, animate and
inanimate things — provided a common speculative framework in which to
fit all these various elements. Ideas taking this as their point of departure and
objective, foreign though they really are to both the aims and the methods of
natural science, have nevertheless had a deep influence on its development;
they have induced a striving after a uniform view of life at periods when
science threatened to disintegrate into aimless detailed research, and they
have produced a love of nature during epochs when humanity had otherwise
turned to abstract philosophic speculation.
The Renaissance produced a number of personalities of this mystical,
half-experimenting, half-brooding type: the Italian Pico de Mirandola,
the Germans Heinrich von Nettesheim — called Cornelius Agrippa — and
Trithemius, and many others. They are, however, of but little interest to
biological history, although their speculations may have in certain cases
indirectly influenced the development of biology during the succeeding era.
One of their contemporaries, however, far more radically than they, furthered
the progress of biology; a man, moreover, who, owing to the life he lived,
is of more than common human interest; namely, Paracelsus.
I3X
0^
<
o
>
H
u
w
0:1
«
<3
c/2
h-I
w
u
<
SEVENTEENTH AND E I G H T E E N T H C E N T U R I E S 133
Theophrastus Hohenheim was born in 1490, or somewhat later, at the
famous monastery for pilgrims at Maria-Einsiedeln in the Canton of Schwyz
in Switzerland. The name Paracelsus, by which he is best known, he him-
self adopted later/ His father, who is said to have been the illegitimate son
of a Knight of St. John of the noble family of Bombast von Hohenheim, was
a physician at the above monastery; his mother, who was probably a peasant
woman, was before her marriage a sick-nurse there. Young Theophrastus,
who was called after the great Athenian botanist and disciple of Aristotle,
grew up in poverty and remained throughout his life, in spite of his ancestors'
nobility, in all respects a child of the people. Nevertheless he received a
sound education, partly from his father and partly from two priests, friends
of the latter; and as a youth he was a student at Basel. However, he soon
wearied of scholasticism, studied alchemy for a time under Trithemius,
referred to above — an abbot who had established a laboratory in his monas-
tery — and afterwards became an apprentice at a mine in the Tyrol, where he
studied metallurgy and was initiated into the professional secrets of the
miners. Even this work, however, did not appeal to him, and he soon joined
the hosts of learned adventurers who ever since the Middle Ages had wan-
dered about Europe under the name of scbolares vagantes or mendicant stu-
dents. Young Hohenheim took his profession more seriously than most of
his colleagues; having wandered through Germany, Spain, and France, he
joined the army with which Christian II conquered Sweden in 15x0, as a
field surgeon. He thus came to Stockholm, proceeded thence to Moscow,
from there again — by which route is not known — to Constantinople,
and finally returned home. In the course of these journeys he naturally had
an opportunity of visiting several universities, but their official learning was
of far less interest to him than the experiences he was able to gain of people
who were at that time believed to be familiar with the occult sciences:
barber-surgeons, witches, gipsies, and executioners. He made such good use
of the medical knowledge thus gained that in 15x6 he was appointed first
town-physician at Basel, with the right to revise the city's pharmacopoeia
and to hold lectures at the University. As a practitioner he was brilliantly
successful with daring cures and simple, cheap medicines, while at the same
time he harassed his colleagues by his extremely overbearing manner. He
gave his course of lectures in German instead of in Latin and started by
ceremoniously burning the classical treatises on medicine, which naturally
increased the hatred of the medical profession. When, moreover, the apothe-
caries, embittered by his sharp criticism, intrigued against him, he had to
1 The name Paracelsus he used as a nom de plume in some of his writings; it probably
means "higher than Celsus" (the Roman physician). Another name which he adopted is Au-
reolus Bombastus, the former meaning "golden," and the latter having been borne by his grand-
father. He also sometimes calls himself "Eremita," after the place of his birth.
134 THE HISTORY OF BIOLOGY
leave Basel after some years. Thus once more he began his wanderer's life,
tramping from town to town in Germany for about ten years, everywhere
winning popularity for his wonderful cures, and hatred for his lack, of con-
sideration and imperiousness towards his colleagues and patients. There
seemed no possibility for him to settle down in peace; rather he had to flee
for his life time and again, until finally Archbishop Ernst invited him to
settle at Salzburg, which he did about the year 1540. Now at last it seemed
that better days were in store for him, but it was not to be for long; in 1541
he suffered a violent death — his calumniators declared it was through an
accident when under the influence of drink, but his friends said it was a
result of a hostile attack. Before his death he had bequeathed his few pos-
sessions to the poor.
Paracelsus^ s -personality
Both in life and after death Paracelsus has been very variously judged: by
some he has been represented as a bare-faced scoundrel, an impudent trader
upon the good faith and superstition of humanity; by others — of early
times as well as in our own day — he has been highly extolled as one of the
boldest spirits that ever lived and one of the greatest promoters of science.
In actual fact, one can find, in his writings and in his life, support for both of
these judgments; on the one hand, uncritical superstition, fantastic paradoxes,
boundless self-conceit, and shamelessly scurrilous language in controversy;
on the other hand, penetrating criticism of his predecessors' theories, and
audacious ideas of his own, aiming far into the future. To brand him as a
conscious cheat would in any event be utterly unjust; rather he is of a type
that has been very common during the Sturm und Drang periods of human
culture. Throughout his writings we come across the same naive self-satisfac-
tion, the same pugnacious temperament, as one finds in several romantic
writers of the beginning of the nineteenth century; like them he was un-
doubtedly induced by an honest intention to "explain the whole of nature
and to reform the whole world." People with such grandiose aims easily
acquire something of the charlatan and humbug in their natures, which,
however, does not exclude the possibility of really splendid traits in their
character. And such Paracelsus certainly possessed; his kindness towards the
poor, his earnest desire to help suffering humanity, his often very poorly
requited loyalty towards his friends, are sufficient proof of that. In his writ-
ings, moreover, he often praises the medical profession as a high and noble
calling, claiming of its members not only knowledge, but also goodness and
morality. Side by side with this there appears in him a feeling of self-respect,
which sometimes finds worthy expression, such as in his motto: ''Nemo sit
alterius qui suus esse potest, ' ' ^ but more often manifests itself as high-sounding
^ "Let no one be another's who can be his own.'"
SEVENTEENTH AND EIGHTEENTH CENTURIES 135
self-praise, as when he says: "My back hair knows more than you and all
your hack writers; my shoe-laces are more learned than your Galen and Avi-
cenna." And his confidence in success is unbounded. "I shall be king and the
kingdom shall be mine . I wish I could protect my bald head from flies
as easily as I do my kingdom, and were Milan as secure against its enemies as
ray kingdom is against you, no Swiss nor Landsknecht would find his way
there." The coarse expressions with which he spices his controversial writ-
ings are such that they could not possibly be quoted. Otherwise his language
is vigorous and original; he wrote in German, like his great contemporary
Luther, for whom in his writings he expresses the greatest admiration.
Alche7nistic conceptiofi of the human body and its junctions
Paracelsus's scientific theories are far more difficult to characterize than his
personality. As the essential part of his activities he always regarded his
medical practice, and his general theories of life invariably have direct
reference to diseases and how to fight against them. And, being originally an
alchemist, at bottom he despised anatomy as he did all detailed research in
general. He sought rather to get the human body and its functions regarded
as a part of the world in its entirety and therefore also as dependent upon the
cosmic process, as it goes on both on the earth's surface and in the firmament
that surrounds it. This was, indeed, just what Aristotle strove after, but
while the latter sought to solve the problem by means of a theory which had
primary reference to the forms of being, to Paracelsus, who from his early
youth had lived in the thought-world of mediaeval alchemy, existence
represented one single mighty chemical process. The alchemistic thought-
structure was, of course, based upon experiments of an essentially magical
character and upon speculations of neo-Platonic-oriental origin, especially
upon the Hebrew cabbala, with its belief in the secret power of words and
graphical signs and their mystical connexion with the things they denoted.
All these elements of learned speculation Paracelsus interwove with all that
he had learnt of folk-magic during his years of wandering, into a natural-
philosophical system of unique character. Its guiding thought probably
emanates directly or indirectly from the cabbala — that is to say, the in-
trinsic connexion that Paracelsus believes is to be found between everything
that exists: the celestial bodies, the things on the earth, and human beings.
This occult connexion, which Paracelsus considers it to be the function of
science to investigate, leads him, the more involved he becomes, into a
mysticism which a modern reader will find extremely difficult to grasp even
in its main features. And the systematic divisions of the subject, with which
Paracelsus is excessively generous, certainly do not make the matter any
clearer. In one of his principal works on science in general, which he himself
published under the title of Paramirum, he divides the causes of sickness,
which he always takes as his starting-point, into five classes: Ens astrale.
136 THE HISTORY OF BIOLOGY
veneni, naturale, spirifuale, deale. What these different ''ens'' really are one
never gets to know; they are apparently mystical powers which produce
diseases and which have different origins. Ens astrale proceeds from the stars,
which have life and can poison the atmosphere, precisely as a person in an
unventilated room pollutes the air in it with his breath. Ens veneni is a cause
of sickness originating in the digestion; each living being has, in fact, his
given food, which is made by that being partly into sustenance for the body
and partly into a poison, which is expelled through the excretive organs;
besides the true excreta, which are the specific poison of the body, quicksilver
is excreted through perspiration, sulphur through the nose, and arsenic
through the ears — the yellow colour of the cerumen in the ear probably
reminded Paracelsus of certain arsenic associations. Thus the ox consumes
grass in its own way and man the flesh of the ox in his. Every being has in its
body an "alchemist," who directs the work; if he gets out of order, the body
becomes sick. In another treatise he is named Archeus, and is apparently to
be regarded as a spiritual being, though no exact description of him is given.
Paracelsus describes in greater detail the third cause of sickness, ens naturale,
and here he expounds his real theory of life and the universe. The human
body is a microcosm, possessing elements corresponding to all the phe-
nomena of the exterior world, particularly to the heavenly bodies; thus the
liver corresponds to Jupiter, the gall-bladder to Mars; the heart is the sun,
the brain the moon, the spleen is Saturn, the lungs Mercury, and the kidneys
Venus. All these organs perform planetary movements in the body, and if
they come into an unfavourable position, disease arises. On the other hand,
they are all independent of food and therefore also of the poisons derived
from it. Moreover, there are included in the body the four elements, as well
as the basic substances of the four temperaments, which are like the gustatory
impressions, sour, sweet, salt, and bitter. All these likewise circulate and give
rise to disease. In regard to ens spirifuale Paracelsus emphasizes the difference
between soul and spirit: the soul is a work of God, but the spirit is created
by the human will and by means of it man can influence his fellow men. Thus
disease can be occasioned by men's hatred; if an enemy makes an image of
wax and maltreats a part of it, his action induces suffering in the correspond-
ing part of the person he desires to persecute — a method to which witches
are particularly partial. Ens deale, finally, is the divine will itself, which gives
sickness and health as may seem good to it; against the divine will medicines
are of no avail, but only piety and prayers.
Paracelsus' s influence
In other writings Paracelsus expounds his theory of the connexion in nature
in different directions; one describes the doctrine of the "signatures" in plants
and their connexion with diseases, as, for instance, that Hypericum, owing
to its perforated leaves, cures wounds from stabs; the peony, owing to its
SEVENTEENTH AND EIGHTEENTH CENTURIES 137
cerebral-shaped pistil, cures paralysis of the brain, etc. To go deeper into
these fantastic ideas is hardly worth while; what has already been stated may
seetn to many readers more than enough of such nonsense. Nevertheless,
Paracelsus 's influence has been both wide and deep. In medicine he was in
many respects a pioneer; he did his best to treat wounds hygienically and
otherwise to leave them in peace; owing to his belief in specific causes of
diseases, he sought for a single cure for each disease, and several of his
methods proceeding on these lines still hold good, especially the use of
mercury in the treatment of syphilis; moreover, this aim of his formed a con-
trast to Galen's and his followers' disastrous attempts to create universal
medicines to be used for all diseases and composed of everything imaginable.
Even modern biology is based, at least in one respect, on a principle laid
dovv^n by Paracelsus; his conception of life-phenomena as fundamentally
chemical processes has without doubt paved the way for modern physiology,
which certainly could not develop before chemical science had been freed
from the primitive mysticism in which it was veiled in the time of Paracelsus
and for a hundred years after him, but which in any case represented a more
stimulating starting-point for the modern idea of substance conversion in the
body than the Aristotelean "cooking" theory, which even Vesalius accepted.
Further, the conception of life as a mystical force uniting the whole of exist-
ence, which Paracelsus, it is true, did not actuafly found, but developed and
stamped with his own original personality, has never, in spite of its many
fallacies and absurdities, succeeded in being entirely suppressed. Time and
again it has been thrust aside by some more exact research based upon actual
facts and referred to the vast public of the dilettanti and mountebanks, but
it never really died out, so that at certain times — for instance, during the
romantic period at the beginning of the last century — it revived with
renewed vigour. And during such periods Paracelsus's reputation has been
freshly enhanced. At all events, history cannot but acknowledge the fertile
genius, the splendid character — in spite of its many exaggerations — and
the force of will with which Paracelsus throughout a life of adversity and
distress fought for what he considered to be the supreme aim of science.
Paracelsus was not very fortunate in his disciples. Cultured people could
not endure his presence for long, and the riff-raff he gathered about him dur-
ing his wanderings was not suitable material for a scientific school. His chief
influence was exerted by his writings, which were read eagerly and produced
a mass of imitations, wherein the defects rather than the merits of the true
Paracelsian writings were conspicuous, and which, published under the
master's name, contributed more than anything else to lower his reputation.
The principal successor to Paracelsus appeared about a generation after his
death, a philosopher who extracted from his writings ana still further dc'
veloped his peculiar conception of life and its functions.
138 THE HISTORY OF BIOLOGY
Jean Baptiste van Helmont was born at Brussels in the year 1577, of a
noble and wealthy family. Left fatherless at an early age, he became a pre-
cocious child, and by his seventeenth year he had already completed his
university studies in philosophy. This, however, failed to satisfy him; he
entered a Jesuit college and there studied theology, especially that of a
mystical character, and this puzzling over life's problems entirely obsessed
his mind. He eagerly studied the neo-Platonists and Paracelsus, for whom
throughout his life he expressed a keen, but by no means uncritical admira-
tion. At the age of twenty-two he took the degree of doctor of medicine,
spent the next few years in travelling through different countries, and then
contracted a wealthy marriage and settled down on an estate in his home
district, dividing his time between scientific research and splendid acts of
benevolence. He carried on his medical practice simply for charity and with-
out any fee; every offer of permanent employment, even the most brilliant,
he firmly declined. He died in 1644.
As already mentioned, van Helmont regarded Paracelsus as his master
and undoubtedly had a certain afhnity with him. True, he possessed none of
his precursor's intrepid geniality, but he was far more cultured, both scientifi-
cally and socially. In his personal character he was mild and lovable, but he
seems to have been of a nervous disposition, which gave his mystical specu-
lations a peculiar quality of exaltation. He had spiritual visions, which he
produced by auto-suggestion through gazing at some strong source of light;
he employed in his researches for obtaining scientific results a direct intro-
spection which he achieved by exalted concentration of thought, both during
the day's work and in the night's rest, when in a state between sleeping and
waking he received inspiration, by which he set great store. This inspiration,
however, often led him widely astray, as when he believed that he had suc-
ceeded in producing rats in a vessel in which some rags and bran had been
kept, or when he imagined he had converted quicksilver into gold, a dis-
covery which so delighted him that he had his son, who was born just at
that time, christened Mercurius.^ But fortunately he had also better inspira-
tions. One of these was his determined opposition to the classical authorities
Aristotle and Galen. Their theories he believed it was his mission in life to
challenge, both because they led to practically worthless results and because
they were pagan. This latter reason is characteristic. While the man of the
Renaissance period, Paracelsus, scorned the classics because they were an-
tiquated authorities and stood in the way of his personal ideas, the emotional
disciple of the Jesuits, van Helmont, felt himself moved first of all to substi-
tute a Christian for the heathen science. This induced him, however, to make
^ This Franz Mercurius van Helmont devoted himself even more exclusively than his
father to purely mystical speculations. The fact of his having published the latter's complete
works constituted his greatest service to science.
SEVENTEENTH AND EIGHTEENTH CENTURIES 139
what in many respects was a sound criticism, particularly of Aristoteleanism,
the weak points of which he brought out very clearly. In his writings he
closely examines Aristotle's theory of the relation of form to reality. Simi-
larly, he rejects the theory of the four elements, for, said he, fire is not an
element at all. Again, when Aristotle believes the fertilizing property of the
sperm to be due to its heat, van Helmont satirically asks how it is that the
cold-blooded fishes are more fertile than any warm-blooded animal. How-
ever, it proves here, as often, to be easier to destroy than to build up; van
Helmont's own theories cannot compete with Aristoteleanism at least in
clearness and consistency. This, indeed, is due partly to the peculiar mystical
principles on which he bases his views, and partly to his lack of stylistic
ability; his writings are extremely obscure and difficult to read. His concep-
tion of nature, like that of Paracelsus, is chemical, with a strong dose of
mysticism, which belonged to chemistry at that time.
Van Helmont' s fermentation theory
The "fermentation process" plays an important part in his theories on nat-
ural phenomena; he had thoroughly studied this phenomenon and had shown
that fermenting must produces a kind of air which is identical with that given
off by burning charcoal and that which sometimes renders the air of caves
irrespirable. For this element he invented the name of gas, a word which has
since been accepted by science. He distinguished several kinds of gas, of
which, however, only the above-mentioned "gas sylvestre," or what we
call carbon dioxide, has been fully described. According to him, digestion
and every kind of conversion of substance in general are due to ferments.
The many different processes of fermentation which he thought he had dis-
covered in the human body were for the most part mere creations of his
fancy, but he had certain ideas which were to prove to be correct, as when
he points out the part played by acid in the cavity of the stomach in the
digestive process, and shows that the undue acidity of the digestive juices is
neutralized by the gall. He was not content with merely establishing facts of
this kind, however; like Paracelsus, he sought to get on the tracks of life
itself and, like him too, saw its innermost essence personified in an archeus,
which is situated in the region of the stomach and controls a number of
subsidiary archei in the other parts of the body. This central archeus, however,
regulates only the material conversion in the body and exists in various
forms in all beings; man has, besides his immortal soul, "intellectus," which
makes the soul participate in blessedness and would wholly control the body
had not the Fall intervened. After the Fall man received a lower soul, "ratio,"
which makes it bound to earth and liable to its impulses, and finally to
death. Beings in the universe have the power of reacting upon one another
by a kind of force operating at a distance which he calls "bias" ; in particular
the bias proceeding from the heavenly bodies has remarkable properties, but
140 THE HISTORY OF BIOLOGY
it would take too long to enter into fuller details. Van Helmont regards
water as the material fundamental element underlying the whole of existence;
out of water arises everything on earth, both inanimate and animate sub-
stance. As a proof of this theory he cites an experiment: he filled a bowl
with loo pounds of dry soil and in it planted a willow, weighing 5 pounds,
and watered it with rain-water. After five years the willow-tree weighed
164 pounds while the soil, when again dried, weighed about loo pounds
less 3 ounces. Thus, he argued, the entire willow-tree was formed of rain-
water. This experiment, which indeed was perfectly correct in both plan and
execution, although the conclusion he drew from it was wrong, testifies
better than anything else that van Helmont was nevertheless a true pioneer
in the field of natural science; to have thought out the first biological ex-
periment based on quantitative calculations is a service to science which well
compensates for many mistakes in the theoretical sphere. Even as a medical
practitioner van Helmont showed the same curious mixture of fancifulness
and foresight: on the one hand he worked at a mysterious universal medicine
which he called "alkahest," while on the other he strongly protested against
the abuses common at that time in connexion with violent blood-letting and
strangely concocted, excessively strong medicines. Moreover, like Paracelsus,
he undoubtedly exercised a strong influence, for both good and evil; his fer-
tile, far-seeing ideas proved of value far into the future, and traces of his
mystical fancies can even be found in scientists of subsequent generations, a
number of whom will be mentioned in the next chapters.
CHAPTER IV
BIOLOGICAL RESEARCH IN THE SEVENTEENTH CENTURY
I. Harvey's Successors
IN THE FOREGOING have bccn described the two entirely contrasted natural
systems which appeared independently in opposition to Aristoteleanism :
the mechanical conception of nature, and the mystical view of life. As has
been shown in the first part of this work, the foundations of the mechanical
view of natural phenomena were laid by Harvey, who proved that the circu-
lation of the blood, which had up to that time been regarded as an expression
for certain vital spirits, goes on as a purely mechanical process. Although
himself a convinced disciple of Aristotle, Harvey thereby laid the foundations
of that modern scientific theory of the phenomena of life which follows the
same methods in dealing with them as those applied to the investigation of
phenomena in inorganic nature. This discovery of Harvey's created an im-
mense sensation; during the immediately succeeding decades after its publica-
tion (in 1 6x8) it was the one great question of the day and occasioned a vast
quantity of literature both for and against it. Its overwhelming truth, how-
ever, soon silenced all opposition; the conservative adherents to the old
system gradually died off and the young research-workers were easily won
over to the new view and devoted themselves to gathering fresh proofs of its
validity. How successful it was is best evidenced by the extraordinary stimu-
lus given to the study of anatomy during the middle and latter half of the
seventeenth century. This period, perhaps more than any other, can be re-
garded as one of brilliant anatomical achievement, to which the preceding
era, beginning with Vesalius's revolutionary inventions in the field of tech-
nique and methods of observation, impresses one mostly as being a period
of introduction. A comparison between these two epochs also produces a
remarkable contrast of a national character; while during the Renaissance
Italy was the sole centre of anatomical research, its range had now spread
northwards: now for the first time England, Holland, and Scandinavia begin
to make definite contributions to the development of biology. And simul-
taneously with this shifting of the centre of biological research we find
another change appearing in its conditions, first in Italy and later the
141
1 42. THE HISTORY OF BIOLOGY
farther north we go. During the sixteenth century the natural philosophers
were still mostly university teachers; such had been both Vesalius and Gali-
leo. In the seventeenth century, on the other hand, and still more so in the
eighteenth century, the universities cease to be the centres of scientific prog-
ress and become instead the seats of unproductive conservatism, mechanically
repeating the formulas inherited from the Middle Ages; the real pioneer
scientists are now private scholars. Descartes, Spinoza, Leibniz, as well as
Harvey and van Helmont, all worked, as we have seen, independently of the
universities, as we shall also find did several of the leading scientists among
their successors, in both the seventeenth and the eighteenth centuries. A new
type of bond of association between men of learning came to be established
in connexion therewith — namely, the scientific societies. Such "academies"
were founded during the seventeenth century all over Europe, earliest in
Italy, afterwards in all countries north of the Alps. Princes and distinguished
people allowed themselves to be nominated as patrons or to be elected honor-
ary members, thereby acquiring an interest in the study of nature. To pro-
mote this study they established laboratories and made collections of natural
objects — so-called "curiosity cabinets" — ■ mostly, it is true, as the name
implies, for their own amusement, but still in many cases for the benefit of
science, owing to the possibilities they offered for research and the grants
of money made by scientists in connexion therwith. All this naturally in-
creased, as it were, the social reputation of science and in this respect offered
a decided contrast to the Renaissance; whereas then the students of nature
had to live in inferior positions, during the seventeenth and eighteenth
centuries many of them held important posts in the community. The period
now to be described was thus in all respects a brilliant one for natural
science — a period which has no counterpart until we come to the latter
half of the nineteenth century.
Discovery of the lymphatic system
As Harvey's immediate successors it is fair to regard those scientists who,
like him, studied the vascular system in man and the higher animals and thus
continued along the path he initiated. Mention has already been made
(Part I, p. 113) of hov/, even during Harvey's lifetime, a hitherto unknown
type of vessel was discovered — namely, the lymphatic duct system — and
how Harvey adopted an attitude of complete ignorance on the subject and
adhered to the ancient tradition. As a matter of fact, a year before the publi-
cation of Harvey's treatise on the circulation of the blood there appeared a
work on "the lacteal veins — a new discovery." — The author, who had
died the previous year, was an Italian physician named Gasparo Aselu
(1581-162.6). He had begun his profession as an army surgeon and afterwards
became for a time professor of anatomy at Pavia, but finally practised in
Milan. There he once carried out, together with some of his colleagues, a
SEVENTEENTH AND EIGHTEENTH CENTURIES 143
vivisection operation on a dog which had just before had a substantial meal,
and thereupon found the peritoneum and intestines covered with a mass of
white threads; he casually cut one of them off and saw that a fluid oozed out;
they were thus not nerves, but vessels of a kind hitherto unknown. Influ-
enced, however, by the traditional Galenian idea of the liver as a blood-
former, Aselli assumed that these vessels, "chyle vessels," as they are now
called, extended from the intestines to the liver. Nevertheless, the discovery
aroused general interest. Aselli 's book was reprinted several times, and a
number of contemporary anatomists interested themselves in this new dis-
covery. Among them there is one who deserves special mention — Johann
Vesling (15 98-1 649), who, though born in Germany, became a professor at
Pavia and in a handbook of anatomy, printed in 1647, gave a detailed ac-
count of the chyle vessels (lacteals). Twenty years after Aselli's work had
been published, however, a young student by the name of Jean Pecquet
(i6iz-74) iTiade a discovery which considerably enhanced the knowledge of
the newly-discovered vascular system. While performing a dissection he
found the canal which forms the common trunk of the lacteals and the
lymphatics — the so-called ductus thorackus. This contradicted Aselli's and
his immediate successors' belief that the lacteals lead from the intestinal
canal to the liver. Pecquet was born in Normandy, studied at Montpellier,
and eventually became body-physician to the minister Fouquet, who was
all-powerful in the early youth of Louis XIV. When Fouquet was sentenced
to imprisonment for fraud, his physician had to go with him, and after that
he disappears from history. He is said to have had a blind confidence in the
power of brandy to cure all manner of diseases — an illusion which rapidly
brought him to ruin. It was instead in Scandinavia, hitherto unknown in
science, that the problem of the lymphatic duct system was finally solved,
and it was a strange coincidence that two scientists from different countries
should have quite simultaneously and independently of each other attained
the same result.
Thomas Bartholin was born in Copenhagen in 161 6. His father, Caspar
Bartholin, was professor of anatomy and a distinguished scientist of the old
school. Having matriculated in his early youth and learnt all that his own
country could teach him, young Thomas at the age of twenty started out on
his travels, which lasted for nine years. First he studied for three years at
Leyden and there became acquainted with Harvey's discoveries, after which
he worked for two years in the anatomical theatre at Pavia, moved on to
Naples, where he was a pupil of the old Severino, then gave a dissertation
for his doctor's degree at Basel, and did not return to his native country
before a professorship was assured to him. As professor of anatomy he did
splendid work, which within a short time made the unknown Danish
university famous throughout Europe; foreign pupils flocked to him, among
144 THE HISTORY OF BIOLOGY
whom may be specially mentioned the German Michael Lyser, who as
prosector used to perform Bartholin's most important dissections, and after-
wards became professor at Leipzig. A number of valuable works were pub-
lished by Bartholin in the sixteen-fifties, but his productive powers very
soon waned. After Lyser's resignation from the post of prosector (1651)
little occurred there in the way of fresh anatomical results; as early as 1660
Bartholin was completely relieved of all instructional and examination work,
and then he had only his membership of the board of the university, of which
he seems to have taken advantage mostly for the pui-pose of providing his
relations with good appointments. With this end in view he is believed to
have passed over deserving pupils, including Steno, who is mentioned later
on. Other not very attractive characteristics of his are also mentioned, such
as that, physician though he was, he once fled from the city during a plague
for fear of infection and retired into the country; again, the way in which
he procured extra salary from the Government caused him to be censured on
several occasions. His writings abound in expressions of the most extravagant
self-praise, though he displays in them a really genuine respect for science —
undoubtedly the most attractive trait in his character. He died in 1680 and
was succeeded by his son Caspar, who carried on his work, reaping a number
of external honours, but in no wise possessing his father's merits.
Bartholin discovers the lymphatic system
Thomas Bartholin's most important achievement is generally considered to
have been his clearing up of the mystery of the lymphatic duct system. He
was not aware of Pecquet 's discovery when he began to study the chyle
vessels (lacteals), and therefore believed, with Aselli, that they led to the
liver. By observation and experiment, however, he soon found that this was
not so and at the same time discovered that these vessels were connected
with a vascular system distributed throughout the entire body, and con-
taining a fluid clear as water. These discoveries he described in a treatise
which came out in 1653, in which he declares that the liver cannot perform
the blood-forming function which classical anatomy had ascribed to it; nor,
since the chyle vessels do not connect with the liver, but on the contrary a
number of lymphatic ducts go in the opposite direction, can the food be
converted into blood in the liver, as the anatomists up to the time of Harvey
and Aselli, and the two latter as well, had concluded. He closes his essay
by erecting a monument to the liver, the body's now dethroned ruler, in the
form of a parody of the high-falutin memorial speeches which it was the
custom of the time to make in honour of the distinguished dead.
Bartholin set great store by his discovery of the lymphatic system and
wrote several fresh articles on the subject, without, however, making any
very important additions to his first account. The rest of his scientific literary
work, which is rather extensive, cannot be compared from the point of view
SEVENTEENTH AND EIGHTEENTH CENTURIES 145
of interest with that dealing with the discovery which he rightly regarded as
his greatest contribution to the development of anatomy. Consequently he
must have been extremely chagrined to find at the very start a competitor for
the honour of having made the great discovery — a youth into the bargain,
without any previous successes to his credit, either scholarly or literary.
Olof Rudbeck was born at Vasteras in 1630, the youngest but one of
eleven children. His father was the imperious Bishop Johannes Rudbeckius,
a man of deep learning, great powers of leadership, and corresponding am-
bition. In his diocese he had founded a college, which, as far as education
was concerned, could compete with the University. It possessed both a
library and a botanical garden. There young Olof received a thorough edu-
cation, and there too he undoubtedly acquired that love for natural science
which induced him, immediately after his entry into the University, to take
up the study of medicine, which was at that time in not very high favour.
Having matriculated at the age of seventeen, he felt himself at once moved to
begin to work on his own account, which the poorly equipped faculty of
medicine rendered it absolutely necessary to do. Like Vesalius, he devoted
himself to the dissecting of animals and in doing so was initiated almost at
once into the then newly-discovered and interesting chyle system. The obser-
vations which he had made within a very short time in this field created such
a sensation that Queen Christina herself desired to become acquainted with
them. In 1651 young Rudbeck was given an opportunity of demonstrating
his experiments before the Queen and as a reward was given a grant to enable
him to travel abroad. Before starting he published his observations in the
form of a dissertation, in the year 1553, and then went to Leyden, where he
studied for three years. When he returned home, he was appointed professor
of anatomy and now devoted himself with great energy to reforming the
system of medical education, which had till then consisted mostly of lectures
on the writings of the ancient authorities. Rudbeck built after his own de-
sign a splendid anatomical theatre, which still exists, and there carried out,
as often as material was available, dissections of human bodies. Exercises of
this kind had never been seen before in Upsala and they consequently aroused
bitter opposition, but Rudbeck did not allow himself to be deterred; on the
contrary, he openly showed his contempt for his opponents' prejudices. To
ridicule them he once caused the remains of a criminal which he had dissected
to be buried with great pomp and drew up for the ceremony a program in
which he delightfully parodied the academical rhetoric of the period. How-
ever, this educational work made too great demands on even his extraordi-
nary energy to allow him time for scientific research; a book on general
animal anatomy which he intended to publish was — undoubtedly to the
immense loss of science — never written. His childhood's interest in botany
bore fruit, for he devoted much of his time to the production of a large
146 THE HISTORY OF BIOLOGY
work composed of botanical engravings and entitled Campi elysii, but this
was never completed , either, owing to the fact that he allowed himself to
be attracted to a new sphere of activity; he became engrossed in that ex-
traordinary, colossal, linguistic-archasological-patriotic work Atland, in
which he seeks to prove that Sweden is the oldest civilized country in the
world. He died in 1701, having shortly before seen a great part of his
scientific production go up in flames in the conflagration which destroyed
Upsala in the same year.
Discovery of the "vasa serosa"
The anatomical work of his youth is, however, all that justifies Olof Rud-
beck's being regarded as the earliest in the long line of eminent naturalists
that Sweden has produced. In his first dissertation in i65z he gives an ac-
count of the circulation of the blood in the true spirit of Harvey and presents
a number of theses, one of which denies the existence of any spirit in the body
other than the animal, while another denies the property of the liver as a
blood-forming organ — proving that he was far in advance of Harvey's
standpoint. In a dissertation printed in the following year he describes
the lymphatic duct system, which he had independently discovered when
attempting to ascertain the structure of the lacteals; he describes the
course of these ''vasa serosa," as he calls them, not only within the trunk,
but also in the extremities; he gives an account of their distensions, the
lymphatic glands; he observes the nature of the lymphatic fluid — that it is
salt to the taste and coagulates in cooking; he tries to ascertain the movement
of the fluid in the vessels by observing their valves; and, finally, he endeav-
ours to work out a theory of the significance which the entire system has for
the body.
Between the two rivals for the honour of having discovered the lym-
phatic duct system in its entirety, Bartholin and Rudbeck, there ensued a
struggle for priority, carried on with the aid of pamphlets, and breaking out
into mutual recriminations of a not very attractive character. National
chauvinism took a part in the game, and the question was debated long aftet
the death of the parties to the dispute, until finally an impartial examination
of the documents put an end to the controversy. After having investigated
the matter R. Tigerstedt came to the conclusion that Rudbeck was the first
to make the discovery, but that Bartholin had the prior claim as regards the
date of publication. That these two eminent scientists both came to the same
result independently of each other would now appear to be beyond all
doubt.
The discovery of the lymphatic duct system constituted a great and im-
portant work supplementing Harvey's discovery of the circulation of the
blood. It was possible now for the first time for research to grapple with the
problem of how the food substances in the digestive canal are utilized by
SEVENTEENTH AND EIGHTEENTH CENTURIES 147
the body, a subject on which the biologists of antiquity had as confused ideas
as they had on the movement of the blood in the veins. Both Rudbeck and
Bartholin, indeed, were prepared to draw the obvious conclusion from their
discovery: since the lacteals did not, as even Aselli believed, lead to the liver,
but just the opposite, it was impossible for that organ to play the all-con-
trolling part in the digestive process which the ancients had ascribed to it.
Both Rudbeck's thesis and Bartholin's funeral eulogy on the liver were con-
firmations of this. It must be admitted, however, that in this both scientists
to a certain extent overshot their mark; research work in recent times has
revealed that a large portion of the substances from the intestinal tube —
carbo-hydrates, albuminous substances, and others — are received by the
ramifications of the cystic vein in the digestive canal and conveyed to the
liver, where they are converted. It would of course be absurd, however, to
ascribe to the opponents of Rudbeck and Bartholin greater prescience in this
respect; conservatism which can give no other reason for adherence to the
past than respect for tradition has no historical justification in face of the
pioneer who bases his ideas on newly-discovered facts, even though he often
overestimates their fundamental value.
As has been mentioned, anatomical research during the middle and close
of the seventeenth century was practised with special keenness in England;
Harvey's achievement acted as a stimulus particularly on his own country-,
men. Space does not allow of a discussion of all the eminent discoverers in
this field of science which England produced during the era in question; one
or two of the most representative will be given here as examples.
Francis Glisson (15 97-1 677) was the son of a landowner; he studied at
Cambridge, first philosophy and afterwards medicine. He took his doctor's
degree in 1634 and as early as two years later became a professor. The Civil
Wars, however, soon compelled him to abandon his educational activities;
he moved to London, where he became a practitioner of repute and one of the
first members of the Royal Society, the scientific association founded in 1660,
which has ever since been the most distinguished centre where the scientists
of England have gathered. Besides some purely medical works, which were
excellent for the period in which they were written, Glisson published two
pioneer works on anatomy, the one dealing with the anatomy of the liver,
the other with the stomach and intestines. The first gives a monographical
account of the liver which, for the then prevailing conditions, was an exemp-
lary work and laid the foundation on which the modern knowledge of the
anatomy of this organ rests. In memory of this, the subperitoneal tissue of the
liver is to this day called Glisson's capsule. But the author does not merely
give a detailed description of the liver; he also expounds a general biologi-
cal theory in connexion therewith, in which he entirely adopts Aristotle's
standpoint. In the component parts of the body he distinguishes matter and
148 THE HISTORY OF BIOLOGY
form and describes them entirely in the Aristotelean spirit: matter, that out
of which something is produced; form, the change in matter which serves a
given purpose; that which produces form is either nature or art. He emphati-
cally protests against the "physical" conception of form and matter which
the "philosophers" advance. His conclusions are formed on the scholastic
model, and the physiological problems that he sets up he solves by a purely
abstract method. Nevertheless, there are among his ideas one or two which
remind one of modern lines of thought, such as his belief that the evacu-
ation of the gall-bladder is caused by nervous irritability. And when it
comes to the lacteals, he shows a knowledge and a comprehension of Pec-
quet's and Bartholin's discoveries. He thus exhibits in his views, as often
happens in transitional periods, a mixture of old and new.
Glisson's younger contemporary and friend Thomas Wharton was a
specialist pure and simple. Born in 161 4, likewise the son of a landowner, he
took the degree of doctor of medicine at Oxford and then practised in Lon-
don, finally becoming head of a hospital there and a highly reputed member
of the College of Physicians. He died in 1673. His fame as an anatomist rests
on his Adenographia, a work in which is given for the first time a comparative
account of the glands of the body. In this book Wharton seeks in the first
place to explain the actual term "gland" and assumes secretion to be the
essential criterion of it; he lays special stress on the difference between the
"viscera," or intestines, and the glands. The tongue is not a gland, but a
muscle; nor is the brain a gland, but a special, "precious" substance. As
belonging to the true glands he characterizes the digestive, lymphatic, and
sexual glands. Wharton discovered the exit of the submaxillary gland — it
now bears his name — and he also for the first time gave a detailed descrip-
tion of the pancreas. The kidneys, the testes, and the thyroids are also care-
fully described. With regard to the glandula fmealis, Wharton denies its
quality of a soul-organ, as maintained by Descartes; he considers it to be an
excretal gland, to which the nerves from the brain drain off waste products,
which are then removed by the blood-vessels — a curious conclusion in
regard to internal secretion, formed more than two centuries before this
process was established in our own time. The hypophysis, according to
Wharton, possesses similar functions, though with a different kind of eject-
ing apparatus. Wharton does little in the way of theoretical speculation; on
the other hand, investigation and discussion of disease conditions in the
glands play an important part in his work.
A more weighty personality we find in Thomas Willis. The son of a
farmer, he was born in i6xi, studied in Oxford, and during that time fought
in the ranks of the Royal Army against the Parliamentary troops. Having
taken his medical degree, he worked as a medical practitioner until, in 1660,
upon the victory of the King's party, he obtained a professorship as a reward
SEVENTEENTH AND EIGHTEENTH CENTURIES 1 49
for his loyal conduct. This appointment, however, he did not retain for long,
but he moved to London and again set up in practice; there he made a great
reputation, became a member of the Royal Society, and published several
treatises. He has been described as an honest man of firm character, towards
the end of his life not very popular at court, owing to his outspoken criticism
of its corrupt morals. He died in 1675.
Willis's principal work is his investigation of the anatomy of the brain
and the nervous system. His exposition of the structure of the brain and of
the nerves proceeding directly therefrom is the first that may be said to ful-
fil modern requirements; to his description of the outward configuration of
these parts posterity has had very little to add. He performed a still further
service to science in having paid special attention not only to the human
brain, but also to that of other vertebrates: in the introduction to his work
he expressly points out that comparative anatomy alone can provide a fully
satisfactory explanation of the structure and functions of the organs. The
account of his investigations into the brains of different animal forms is
illustrated with very fine engravings, a number of which were drawn by his
friend, the famous architect. Sir Christopher Wren, the designer of St.
Paul's Cathedral in London. With regard to the functions of the nervous
system, Willis, in contrast to the Aristoteleans Harvey and Glisson, asso-
ciated himself entirely with the theory advanced by Descartes. The manifesta-
tions of life are induced by currents in the nervous system, which penetrate
the brain and according to their nature are distributed into the different parts
thereof; the world of ideas and the memory he places in the cortex of the
great brain. Willis thus forestalls Swedenborg's radical investigations into
the localizations in the brain, which are certainly considerably deeper and
more detailed than his precursor's. Willis also carefully studied the func-
tional spheres of the different nerves; thus, by binding up the vagus nerve in
a live dog, he established their influence on the lungs and heart.
In a work published at a later date Willis deals with the soul of animals.
This work has a far wider range than its title implies, for it contains a mass
of information of various kinds. Thus, he gives an account of a number of
investigations he made into the anatom^y of invertebrates, which would
have been of interest to his age had not at the same time Malpighi and
Swammerdam given far better accounts. The main purpose of the work, how-
ever, is a comprehensive study of the vegetative and sensitive soul, which in
his view — and in that of Descartes — is common to the animals and man.
Man has, indeed, his rational soul as well, which is immaterial and can there-
fore survive after death; the soul under discussion here is, on the other hand,
that material vital spirit which finds expression in currents in the nervous
system and which produces all the manifestations of life in animals and those
that are purely animal in man. How it happens that animals can in some cases
150 THE HISTORY OF BIOLOGY
perform actions which indicate a conscious intelligence is a problem which
causes the author considerable difficulty; true reason cannot exist in animals,
for then they would soon come to resemble man and would, moreover, be
immortal; it must therefore be the material vital spirit that, either instinc-
tively or mechanically, directs their actions. In endeavouring to apply this
theory Willis becomes involved in a maze of speculations that it would be
hopeless, and indeed would take too long, to follow in all the subtleties to
which they give rise; the numerous learned authorities whom he quotes
merely give additional confirmation of the state of helpless confusion into
which psycho-physiological speculation had drifted ever since it left behind
it the safe harbour of Aristoteleanism. When a scientist of Willis's rank can
seriously discuss the question whether the vital spirit can be compared with
spiritus vini or hartshorn oil, it is easy to realize to what hopeless lengths
the natural-scientific speculation of that age could go. Nevertheless, there
were even at that time students of nature who applied with far more sub-
stantial results the newly-discovered exact method of research in the sphere
of biology. Examples will be given in the next section of this chapter; first,
however, we may give one more example of radical anatomical research dur-
ing this epoch.
Raymond Vieussens was born in 1641, of a military family; he devoted
himself to medical studies at Montpellier and after having graduated became
a doctor at a hospital there. He paid specially keen attention to the study of
the structure of the nervous system and after many years of preparatory work
published his great Neurologis universalis in 1685. This brought him immediate
fame. He was summoned to the court in Paris and was for a time physician
there, but afterwards returned to his old post, which he retained until his
death, in 171 5. His description of the nervous system is remarkable for its
unprecedented accuracy and completeness in its anatomical details; for us it
is of special interest as representing the foundation upon which Swedenborg
based his ingenious speculations upon the connexions of the nervous system,
which Vieussens studied and illustrated with great exactitude. Unfortu-
nately he, too, became involved, with but little success, in physiological
speculations upon the "spiritu/' of the nervous system, which he believes to
be secreted in the brain from the blood circulating through it to the latter,
as well as upon a "spirifus nitro-aerius," contained in the blood itself; a kind
of acid component thereof, which is drawn from the air and from the con-
stituents of food. As a result of these ideas he became involved in a tedious
controversy. He also brought out a couple of anatomical works on the heart
and the vascular system, but they are of less value than his neurology.
SEVENTEENTH AND EIGHTEENTH CENTURIES 151
i. Attempts at a Mechanical Explanation of Life -phenomena
Giovanni Alfonso Borelli was born at Naples in 1608. His father had been
an officer in the service of Spain. At an early age young Alfonso showed a
marked genius for mathematics; in order to complete his education in this
subject he went to the University of Pisa. There Galileo had once been pro-
fessor and he now lived, an honoured and influential man, as court astrono-
mer in the neighbouring city of Florence. It was not surprising, therefore,
that Borelli was won over to his physical and astronomical theories, and he
entered with ardour into the new field of research which they opened up for
him. After a period of deep study he was elected professor at the University
of Messina by the Government of his country, and there he gave instruction
for some years. Conditions at that university, however, were limited and
offered him no scope, wherefore Borelli returned in 1656 to Pisa, whence he
was summoned to Florence in the following year. In the latter city the disci-
ples of Galileo had founded a free academy, called "Accademia del cimento";
here Borelli found employment and worked for ten years. He went in seri-
ously for medical studies with a view to applying Galileo's physical princi-
ples to medicine. Unfortunately for him, however, he was induced by the
promise of a higher salary to return to Messina. As a matter of fact, shortly
afterwards, in 1674, ^^^ inhabitants of that city formed a conspiracy against
Spanish rule in Sicily. The rebellion having been quelled, Borelli, who had
leagued himself with his countrymen, had to save himself by flight. Ruined,
the already ageing philosopher arrived at Rome and there at first obtained
employment in the service of Queen Christina of Sweden, who in her exile,
as once she had done in her native land, loved to surround herself with distin-
guished scientists. For some years he remained her private physician and
actually dedicated the best of his writings to her. But a fresh misfortune
befell him; through the dishonesty of a subordinate he was again ruined,
and the Queen, whose own aff'airs were, as is well known, in a bad way, was
unable to assist him. He had to take refuge in a monastery, and there he died,
in 1679.
The movements of animals
BoRELLi's restless life may possibly have been due to his temperament, which
is said to have been reserved and morose. However, he enjoyed a universal
reputation as one of the foremost scientists of his time, and his literary
production was of an exceptionally many-sided character. He carried on his .
teacher Galileo's work in physics and astronomy. But his books on these
subjects, however valuable they may have been, were entirely overshadowed
by his great biological work, De motu ani?nalium, which was published in the
same year in which he died, dedicated, as mentioned above, to Queen Chris-
tina, who, according to the preface, defrayed the cost of printing. Through
152. THE HISTORY OF BIOLOGY
this work he ranks by the side of Harvey as one of the leading pioneers of
modern biology.
"As is generally done in other physical-mathematical sciences, we shall
endeavour, with phenomena as our foundation, to expound this science of
the movements of animals; and seeing that muscles are the principal organs
of animal motion, we must first examine their structure, parts, and visible
action." In these words Borelli states his views on the function of biology
and thereby declares his starting-point to be Galileo's conception of nature,
and the work reveals the fact that here, as with Galileo, we are face to face
with "a new science." Even the very arrangement is original: by means of
short sentences, which, on the model of Euclid, are called propositions, with
accompanying proofs and corollaries, the inquiry is led from the simplest
element of the motory system, the individual muscle, gradually to more and
more complicated organs and organic systems, until finally the whole of the
being's power of movement is described in the form of a summary. First of
all, of course, he discusses the movements in man, to whom the lion's share
of the work is devoted, after which he studies the movements of mammals,
the flight of birds, the swimming of fishes, and even the characteristic move-
ments of insects and others of the lower animals. The first propositions are
introduced by an analysis of the actual muscular substance, Borelli maintain-
ing that these elements of bodily movement are identical with the flesh —
which the Aristoteleans denied. Then he describes in detail the different
mechanical functions of the muscles, this being clearly explained in schematic
form, with figures appended. On this basis he then proceeds to a study of the
different forms of movement: first the individual extremities, then the move-
ment of the whole body under different kinds of action, lifting, walking,
running, jumping; and even, what was for a southerner an unusual form of
motion, that of skating is observed and analysed. After the movements of
mammals, as above mentioned, have been examined, he describes birds'
power of flight in comparison with the foregoing movements, and finally he
analyses the action of swimming, in which he pays attention not only to the
motory system of fishes, but also to the possibilities of man and other land-
creatures in this respect. In connexion herewith Borelli puts forward sug-
gestions for a diver's dress and a submarine vessel; whether he was ever in a
position to make practical tests of these inventions history does not relate.
M-Uscular -physiology
In another part of his work Borelli seeks to explain the causes of muscular
action. For this purpose he first of all tries a number of purely mechani-
cal alternatives, which he rejects, among them being the possibility of the
muscle's being shortened merely by contracting its mass, produced by the
concentration of its smallest particles. Such takes place, according to Borelli,
when a piece of red-hot wire becomes shorter on cooling, but this cannot be
SEVENTEENTH AND EIGHTEENTH CENTURIES 153
assumed to take place in muscular contraction, which, indeed, occurs as the
result of impulse from the nervous system — that is to say, through fluid
flowing to it from without. Here obviously comes in Descartes's theory of
the currents in the nerves; on the basis of this theory Borelli concludes that
the swelling of the muscles upon contraction is caused by a process of fer-
mentation, which arises when the blood in the muscles is mixed with the
nervous fluid flowing into it. For the fact that Borelli was unable, with the
means available at that time, to explain the extremely involved co-operation
of physical and chemical processes which constitute muscular contraction,
he cannot, however, be blamed; on the contrary, his assumption that a
complicated chemical action and not a simple mechanical one is here in-
volved must be admitted to be an inspired presentiment of what our modern
science has at last definitely established. After having then given an account
of a thorough investigation of the muscular mechanism of the heart and the
respiratory system, Borelli concludes his work with some speculations on
the subjects of digestion and fertilization, which are partly based upon the
opinions of his precursors and are otherwise not very successful, so that they
may be passed over here. The same is true of his purely medical speculations,
such as his theory that fevers do not originate in the blood, but in the nerv-
ous fluid, and other assumptions in connexion therewith.
Borelli creates experimental biology
Borelli was above all a mechanician and his greatness lies in his having
created experimental biology operating with purely mechanical forces. In
the introduction to his work, it is true, he gives the assurance that all the
mechanical phenomena in the living body, which he proceeds to describe,
are produced by the vital spirit — had he not admitted this, he would cer-
tainly never have obtained the papal authority to publish his work, which
now adorns the first page of his book — but having once made this theoreti-
cal reservation, he carefully avoids in the discussion that follows the inclusion
of any other points than the purely mechanical. And it is just for this reason
that, in spite of occasional weaknesses, his work stands out as the first to
apply all through the fundamental principle on which modern biology is
based.
Borelli was highly appreciated by his own and the succeeding age; thus,
the great Dutch physician Boerhaave advises every doctor to read the work
De motu animalium. Although he was certainly the foremost, he was not
the only scientist of his kind to tackle biological problems from a purely me-
chanical point of view. Of those of his contemporaries who distinguished
themselves in this respect two in particular are worthy of a detailed account.
Claude Perrault was born in Paris in 161 3, the son of a lawyer. He
studied at the university there; first of all, mathematics and the classical
languages, and then chemistry. Having taken a doctor's degree he practised
154 THE HISTORY OF BIOLOGY
for a time, but became more and more attracted to architecture. It is in this
sphere that he became best known: as the designer of the Louvre colonnade
he is mentioned in every guide-book to Paris. How^ever, the interest in
anatomy that he acquired as the result of his medical studies he maintained
throughout his life; he dissected animals of every available kind and com-
pared the results he achieved. Finally he fell a victim to his own zeal: he
died (in 1688) of blood-poisoning contracted when dissecting a camel that
had died in the Royal Zoological Garden.
The work in which Perrault recorded his biological speculations bears
the characteristic title of Essais de la physique. The third of the four volumes
that the work comprises is called Mkbanique des animaux, and in it he has
developed his ideas on the functions of animal life. The work was published
in 1680 — that is to say, at the same time as Borelli's, and, of course, quite
independently of the latter's. As already mentioned, Borelli was a disciple
of Galileo; Perrault, on the other hand, shows himself in his writings to be
manifestly influenced by Gassendi, although accessible biographies make no
mention of either any personal or any literary contact between them. Gas-
sendi based his conception of nature on the ancient atomic theory, such as it
has been preserved in literature, mainly through Lucretius; moreover, he was
an admirer of Galileo and an opponent of Descartes. In Perrault we find views
on all these questions in full accord with Gassendi. In his first chapter he
states matter to be composed of individual particles, at the same time hard
and elastic; the air in particular is composed partly of finer, spherical and
partly of coarser, cubiform particles. On the other hand, when it comes to a
question of gravity, Perrault shows himself familiar with Galileo's discover-
ies in that field, and finally he sharply criticizes Descartes, particularly his
theory that animals lack consciousness. In opposition to this assertion
Perrault maintains the independent and peculiar intelligence of animals,
citing numerous examples.
Perrault' s philosophical method
With regard to the knowledge of animals, as also with regard to physics,
Perrault lays down two scientific methods: the "historical," which is purely
descriptive, and the "philosophical," which seeks to ascertain the causes
of what takes place in nature. Following this philosophical method, Per-
rault deals with special phenomena in the animal kingdom, wherein he
endeavours always to find out the mechanical connexion of causes, declaring
that this is all that it is possible for m.an to discover. He entirely disagrees
with both the older, idealistic philosophy, which scorns to have anything
to do with natural phenomena, and the younger philosophy — that is, that
of Descartes — which denies all manifestations of soul in animals. Perrault
then tries to ascertain the mechanical course in quite a number of vital func-
tions; particularly sense-impressions, the digestion, and the external move-
SEVENTEENTH AND EIGHTEENTH CENTURIES 155
merits of the body. On all these subjects he expresses original opinions,
constantly pointing out comparisons with mechanical organizations in ani-
mate nature, and in doing so he displays throughout his capacity as a prac-
tical technician. Thus, he compares the muscles which act counter to one
another in an arm with the shrouds on a boat-mast which counterbalance
one another; in another connexion the valves of the heart are compared with
the mechanism of a sluice-gate. His study of the mechanism of the auditory
organ is full of striking observations; in regard to the visual organ, too,
he makes some interesting comparisons — for instance, between the lenses
in different animals — although naturally the results of modern optics,
founded by Huygens and Newton, are unknown to him. In every case he
tries in the first place to ascertain the nature of the movements performed
by the different parts of the body; consequently he pays special attention to
the structure and function of the muscles; in contrast to Borelli, however,
he entertains the false idea that it is not "the flesh," but the interposed
lengthwise and crosswise fibres — that is to say, the connective tissue —
which expands and contracts. He gives a detailed description of this proc-
ess; he believes that the muscle in its natural position is contracted; when
it relaxes, it does so because the nerves convey to the muscular fibres a
"substance sfiritueuse,'" which expands them just as metals are expanded by
heat. The motory impulses therefor come from the brain; of its parts he con-
siders the medulla oblongata to be the most essential and the great brain the
least important; he had, indeed, observed that it is possible to remove the
great brain from a live dog without the animal's dying, but if the medulla
oblongata is injured the dog dies at once. This fact the physiology of our own
day has, of course, confirmed, although the conclusion which Perrault draws
from it as to the lesser importance of the great brain is wrong. To "peris-
taltics," by which he means the movements within the body, he devotes
a special chapter, which likewise contains many keen-sighted observations,
the process of nutrition as a w^hole being of particular interest to him; natu-
rally, however, he has a number of false ideas as to its chemistry, such as
his belief that the air contains directly alimentary constituents, proved by
the fact that some broods of serpents, which were kept in a jar without
food, developed "on air." Owing to the limited possibilities of investi-
gation in those days, we cannot blame him for being unable to discover that
animals have an embryonic reserve nutriment to live on.
In spite of occasional fallacies, Perrault may thus be regarded, side by
side with Borelli, as one of the pioneers of modern biology. Besides these,
the Danish philosopher Steno, well known for the strange fate that befell
him, is worthy of mention as having been active in the same direction.
Nils Steensen, known under the latinized form of his name, Nicolaus
Steno, was born in Copenhagen in 1638, of a wealthy family of goldsmiths.
156 THE HISTORY OF BIOLOGY
He early showed an inclination for medical studies, which he was able to
carry out first at home under Bartholin, and then in Amsterdam and Leyden.
He soon made important anatomical discoveries, which gained him a Euro-
pean reputation. After this he spent some years in his native city seeking
employment at the University, but as he was constantly passed over in favour
of the relations of Bartholin, he grew weary of waiting, and having received
his paternal inheritance, he journeyed to Paris. There he studied cerebral
anatomy for a time, publishing a treatise on that subject, and then went to
Italy, where he worked at Pavia and Florence. The Grand Duke of Tuscany
was very gracious to him and gave him money to continue his studies and
his publications. Under these new conditions, however, Steno underwent
a severe spiritual crisis, resulting in his being converted to Catholicism, a
step which brought him brilliant worldly advantages, but soon completely
upset his intellectual balance. After spending some time in his native country,
where the authorities had now learnt — too late — to estimate him at his
true value and offered him a position that would bring in a good income,
though at the same time they naturally did not look with favour upon his
religious conversion, he returned to Italy, took holy orders, and was soon
appointed bishop and chief organizer of Catholic propaganda in north Ger-
many. In the latter capacity he displayed fanatic zeal; he addressed a letter
to Spinoza, amongst others, with whom he had been acquainted in his
youth, urging him to become converted, which the latter declined to answer.
At the same time he gave himself up to violent asceticism, which rapidly
undermined his health. He was only forty-eight when he died (in the year
1686), and he was buried with great pomp at Florence, where a fine monu-
ment in the Church of San Lorenzo perpetuates his memory.
As an anatomist Steno devoted himself principally to the study of two
organic systems: the glandular and the muscular systems. With regard to
the glands, Glisson and Wharton had, of course, been the pioneers, but Steno
at any rate made fresh and important contributions to the knowledge of
these organs; he discovers the exit of the parotid gland, which has been
given the name of " diictus stenoniatius" ; he thoroughly explained the anat-
omy of the other glands of the mouth, and, lastly, found the exit of the
tear gland. As an anatomist of the muscular system his chief aim was, as
he himself says, to apply to anatomy the laws of mathematics and thus to
create a geometrical system for the muscles. His main work on muscular
investigation, in which he carefully analyses, along the lines just indicated,
a number of muscles, starting from their simple component parts, was pub-
lished twelve years before Borelli's important work. In this book Steno's
theory is expounded, though Borelli expressly states that its rules apply only
in certain special cases. In actual fact Steno's work deals with a far more
limited field of investigation than Borelli's; in method it is more speculative
SEVENTEENTH AND EIGHTEENTH CENTURIES 157
and less experimental than the latter's, and therefore, though published
earlier, it has not the same universal application as the work De motu animal-
ium.
Steno as a paleontologist
Besides the above work Steno found time during the short period which he
devoted to natural science to make a number of important contributions.
Thus, he discovered the ovaries in the shark, which, as is well known,
produces its young alive — a discovery the importance of which he himself
fully realized; up to that time people had thought that ovaries existed only
in oviparous animals. Steno's greatest service to biology, however, is his
creation of modern palaeontology. Already during his first visit to Florence
he had had an opportunity of studying a kind of stone images found in great
numbers there, which the inhabitants called " glossopetrf or stone-tongues,
and by means of comparative study he proved that they must have been the
teeth of sharks. During his later sojourn in Tuscany he carried out a system-
atic study of that district's geological strata, and thought himself justified
in concluding from their position and appearance that they had been strati-
fied out of water, a fact which he believed to be still further confirmed by
the quantity of animals and plants found in them. Supported by these facts,
he outlines a theory of the origin of the earth's strata which is a presage of
present-day geological science. He never got further than this rough out-
line, however; the results of his geological investigations would not, as
the times were then, accord with the Church doctrine that he had so zeal-
ously embraced, and, indeed, this was one of the reasons why he completely
abandoned a science which he had initiated and studied with such splendid
results. A tragic fate indeed, although by no means without its counterpart
in scientific history. In this his last natural-scientific work, however, Steno
deals also with other problems besides the purely geological and pateonto-
logical. His geological stratification theory is really only one link in a gen-
eral theory regarding transubstantiation in nature, according to which all
things that exist have originally been and are still being precipitated out of
fluids. He thus comes to the question of the crystallization process in the
mineral kingdom, which he investigated with great thoroughness and good
results, and in connexion therewith he discusses the transubstantiation and
organic formation in animals, which he likewise conceives to be a strati-
ficational process similar to that which takes place in inanimate nature, a
precipitation from fluids, of which he distinguishes various kinds in the ani-
mal and plant organism. Thus he reconciles the changes in substance in ani-
mate and inanimate nature under the same point of view and without giving
any idea of the essential diff^erence between the growth of a crystal and that
of a living organism. This, then, clearly denotes the limitation in the me-
chanical conception of nature in the seventeenth century — a limitation
158 THE HISTORY OF BIOLOGY
which undoubtedly contributed to the fact that the promising ideas which
such investigators as Borelli, Perrault, and Steno had produced were not
followed up. The insignificant progress made in the sphere of organic chem-
istry at that period in fact rendered impossible the expansion of experimental
biology beyond the purely mechanical sphere in which the scientists here
mentioned achieved such splendid results. Another reason why biological
research took a new direction, however, proved of still more decisive im-
portance — the discovery of the microscope, and its constant improvement,
resulting in the opening up of hitherto unguessed possibilities for biology.
We shall now proceed to discuss this method and its representatives.
3. Microscopies and Microtechnology
The fact that ground lenses magnify the vision seems to have been
already established in classical antiquity. Eye-glasses and simple magnifying-
glasses came into use in the sixteenth century; the inventors of complex len-
ticular systems are commonly said to have been two Dutch spectacle-makers,
Janssen by name, father and son. These earliest microscopes must have been
extremely primitive: a tube with a plate for the object, without any adjust-
ing apparatus, and the lens or lenses at the other end of the tube; the tube
was held to the eye and directed when in use towards the light like a tele-
scope. The magnification was, to start with, not more than ten times, but
it nevertheless excited general wonder, especially when tiny live creeping
things were put under the microscope and could show their movements. It
was considered particularly fascinating to watch fleas, from which the earli-
est type of microscope received the name of " vitrum pulkare," or flea-glass.
During the seventeenth century, however, the construction of the micro-
scope, chiefly the system of lenses, was considerably improved, with the
result that good individual instruments made by clever masters in the art,
such as the Dutchman Leeuwenhoek, mentioned below, could magnify up
to iyo times. But throughout the eighteenth century and for a good part of
the nineteenth, microscope construction underwent but few changes, except
for isolated improvements, such as the introduction of a stand and mirror,
and it was not until the thirties that the long line of new inventions that
have gradually made the microscope what it is today had their beginning.
Microscopy has had, therefore, two periods of brilliant achievement in the
course of its history: the seventeenth century, and the latter half of the nine-
teenth century up to the present day.
Even Harvey, according to his own statement, used a " perspicillum"
when studying the circulation of the blood in insects. The first scientific
treatise that is based exclusively on microscopical investigations was the
SEVENTEENTH AND EIGHTEENTH CENTURIES 159
Italian Francisco Stelluti's study of the structure of the bee, which was
published in Rome in 16x5. But foremost among those who systematically-
based their research on magnifying apparatus must be mentioned the Italian
Malpighi.
Marcello Malpighi was born in i6i8 at Cavalcuorc, a place near Bo-
logna, where his father owned an estate. Here Marcello spent his childhood.
At an early age he became a student at Bologna and devoted himself to the
study of Aristotelean philosophy. He had to break off his studies, however,
upon the death of both his parents, in 1649, ^^'^ ^^'^ ^^ leave the University
for a year or two in order to settle his father's affairs and look after his
younger brothers and sisters. With this latter end in view he returned to the
University, after a short time graduated as a doctor (in 1653), and then
devoted himself to medical practice and university teaching. His brilliant
gifts were soon apparent, but certain intrigues delayed his advancement, and
when at last — in 1656 — the Senate of Bologna instituted a professorship
for his special benefit, he preferred to accept an appointment to a chair of
medicine at Pisa. There he got to know Borelli, and their acquaintance de-
veloped into a lifelong, firm friendship, Borelli in the beginning possessed
a tutorial influence over his colleague, who was twenty years younger, and
taught him to realize the defects in the Aristoteleanism which he had till
then embraced. Malpighi, however, considering that the climate of Pisa was
bad for his health, returned once more to Bologna, but shortly afterwards
he was called, upon Borelli's recommendation, to be professor at Messina,
at a good salary. After four years, however, he relinquished this appoint-
ment, owing to intrigues and troublesome interference on the part of the
authorities. So for the third time — in 1666 — he returned to Bologna,
where a professorship awaited him, which he held with honour for twenty-
five years. In 1691, being then in his sixty-fourth year and in failing health,
he went to Rome and became private physician to the Pope, and he died
there of apoplexy three years later — in 1694.
In contrast to most of the biologists of the earlier period, but like so
many of those of the present day, Malpighi published his observations not
in large consecutive works, but in the form of short reports, sometimes com-
prising only a few pages, usually sent in to the Royal Society of London,
of which he was a member and which undertook the printing of them. Prac-
tically every one of these small papers contained some important discovery
in different branches of biology. The connecting link in this literary work
is not any common idea running through it all, but is represented by micro-
scopical technology, which Malpighi with hitherto unrivalled genius ap-
plied to every imaginable object in living nature that came within his range.
Thus Malpighi was the founder of microscopical anatomy in both the ;:nimal
and the vegetable kingdoms. One reason for the disconnected way in which
l6o THE HISTORY OF BIOLOGY
he recorded his experiences may perhaps be found in his lack of capacity
as a writer; indeed, from the point of view of style his papers were not of
a high standard, his exposition being often unclear and at times almost
impossible to understand.
Malpigbi's investigations
Of Malpighi's writings the first in point of time and undoubtedly the most
important as to contents is the short account, in the form of two letters
addressed to Borelli, of his investigation of the structure of the lungs. In
the first of these essays he declares that the substance of the lungs had till
that time been regarded as "fleshlike," which was incorrect; the lung con-
sists rather of a network of extremely thin-walled cells, which are connected
with the finest ramifications of the windpipe. This, he states, can best be
observed by flushing out with water the blood from a fresh lung, then in-
flating the lung through the windpipe, and afterwards drying it. In con-
nexion with this discovery he advances some speculations with regard to
the function of the lungs, which he assumes to be to keep the blood flowing
and to prevent it from coagulating, which happens when it has run out of
the veins. He also discusses the high temperature which fever produces in
the blood and considers it to be due to a process of fermentation. In the
second letter he gives an account of the finer structure of the lung of the
frog, and in connexion therewith he describes his discovery of the capillary
circulation as a connecting link between arteries and veins, which he also
observed in the frog. In order to demonstrate this vascular system he recom-
mends that a frog's lung be inflated, then dried, and in that state examined
under a magnifying-glass. He himself emphasizes the importance of the fact
that the transition between the venous and the arterial blood had been dis-
covered, and posterity has confirmed the truth of his discovery. The achieve-
ment that comes next in importance is his investigation of a series of organs
which he placed in the category of the glands. These investigations he carried
out partly with fresh material, partly with such as had been hardened by
cooking, besides which, by means of injections into the blood-vessels and
the preparation of the tissues, he endeavoured to trace the minutest elements
of the organ. In the liver, with which he started his investigation, he thus
followed the blood-vessels up to their finest ramifications, which he con-
nected with a mass of small protuberances which may be brought up on
the cooked liver. By establishing the existence of these he considered the
liver's glandular character proved, which modern science has shown to be
correct, although the small protuberances are actually pure outgrowths
without any equivalent in the true structure of the liver. Malpighi also in-
cludes in his investigations of glands his observations of the cortex of the
cerebrum. He observed in this organ the pyramid-cells, which he believes
to be glandular elements that secrete the ' fluidum" whereby the muscles are
SEVENTEENTH AND EIGHTEENTH CENTURIES l6l
moved to contract. The nerves, which are hollow, form the passages out-
wards for this fluid, the nature of which Malpighi does not further describe,
although, like Willis, he seems to regard it mostly as some kind of fugitive
liquid. For the rest, he has made contributions to the knowledge of the
blood-vessels' ramifications in the brain; on the other hand, his speculations
in regard to the function of the cerebral cortex are not very enlightening,
just as, on the whole, Malpighi was more of a practical observer than a
theoretician. It was left to Swedenborg, a couple of generations after Mal-
pighi, to work out an explanation — partly based on the latter's observa-
tions — of the localizations in the cortex of the brain, which even our own
age might well think remarkable. — Finally, Malpighi studied the kidney
and the spleen, using the same methods as those applied to his observations
of the above-mentioned organs, and in this field, too, he achieved valuable
results; in the kidney he established the course of the blood-vessels and of
the tubules and has in general given a good description of the inner structure
of the organ in man and in several other mammal forms; the glomeruli of
the kidney still bear his name, and likewise the name of the Malpighian
follicular bodies in the spleen testify to his powers of observation. Extremely
useful has been Malpighi's monograph on the tongue, the muscles and nerves
of which he explained and the papillas of which he described and charac-
terized as gustatory organs. And finally he published an account of the de-
velopment of the hen's egg, which forms a creditable supplement to the
investigations previously carried out by Fabrizio and Harvey. In the sphere
of invertebrate biology Malpighi has also performed a service by investi-
gating the structure and history of the development of the silk-worm; he
discovered in this subject the excretal organs characteristic of the Tracheata,
which are now called the Malpighian tubes, and in other respects, too, he
laid the foundations of our knowledge of the anatomy of insect larval forms
and likewise made valuable observations regarding the butterfly's evolution
out of the pupa and its anatomical structure.
Malpighi's works on vegetable anatomy
There still remains to give an account of Malpighi's activities as a pioneer
in a quite new field — vegetable anatomy. Biology, as a universal science
of life and its manifestations, has for obvious reasons been based principally
on the study of those creatures which have stood in the closest relation to
man — that is to say, first and foremost man himself, and secondly the higher
and lower animals; for the purposes of this science plants have as a rule come
last. There are two fields of biological study, however, in which plants have
from the beginning been more useful for standardizing purposes than ani-
mals — namely, classification and the cell and tissue principle. The fact that
plants have proved a more convenient starting-point in this latter sphere
is, of course, due to their having, on account of their cellulose formations.
l6l THE HISTORY OF BIOLOGY
such extremely easily discernible elements. Thanks to the resultant structural
conditions, which in their main features are distinguishable even to the
naked eye, plants have constituted the starting-point for the study of the
elementary nature of living matter as a whole. And the honour of having
introduced this study into science is due to Malpighi, even though he may
have had to share it to a certain extent with another, the English physician
Grew.
The results of Malpighi's investigations into the subject of vegetable
anatomy were, after ten years of preparation, submitted to the Royal Society
of London and were there published. They consist of a comparative study of
the anatomy of different plants, both ligneous plants and herbs. First the
structure of the bark is described, then that of the wood and pith, and finally
the buds, leaves, flowers, and fruits. The different parts of these plants are
composed of small "utriculi" or cells, which can be distinguished by means
of a magnifying-glass and which in their turn form a larger connective group.
The cuticle and bast of the bark, the vesicular system of the wood and its
fibres are analysed, special interest being devoted to the spiral vessels, whose
inner spiral thickening induces a comparison with the tracheal system of
insects, in regard not only to structure, but also to function. Upon this chance
similarity Malpighi now bases a universal theory of respiration applicable
to all living creatures — which, for all its conjectural ideas, represents a
shrewd guess as to the uniformity of life-phenomena in all organisms. He
believes that the more perfect the living beings are, the smaller their re-
spiratory organs are: man and the higher animals do with a pair of lungs
of comparatively small size, whereas fishes have numerous closely ramified
gills, and the tracheae of insects spread throughout the entire body, while
again the spiral vessels in plants develop in such quantities that they fill
up even the most insignificant ramifications of the individual plant. Plants,
he supposes, take up air out of the soil through the roots; the leaves possess
no openings that could serve this purpose. With regard to the significance
of respiration for living beings, he believes that it consists in promoting the
mobility and "fermentation" of the alimental juices. On the whole, fermen-
tation plays much the same part in Malpighi's physiological speculations
as "cooking" does in Aristotle's; at any rate, it constitutes an advance, in
spite of the indefiniteness of the idea, which was inevitable considering the
stage of development which chemistry had reached in those days. In con-
nexion with his account of the elementary constituents of plants Malpighi
advances a number of general physiological speculations, all intending to
demonstrate similarities between vegetable and animal organisms and their
functions. In doing so he only follows, it is true, a principle which was pecul-
iar to a botanist of that time and which had its origin in Cesalpino's phys-
iological speculations on plants, to which we shall revert later on. It is,
SEVENTEENTH AND EIGHTEENTH CENTURIES 1 63
however, but natural in the circumstances that these comparisons should
lead to false conclusions, and, as a matter of fact, they did to a great extent
prevent Malpighi from taking advantage of the promising material for study,
which otherwise he might possibly have been able to do. Thus he compares
the buds, out of which gradually sprout leaves and branches, with the ovary
and the uterus; then he deals with the flowers and carefully compares their
special parts in different plants; as he fails to clear up the question of their
sexuality he advances the theory that the flowers serve to purify the juices
of the plants before germination, just as menstruation precedes pregnancy.
He studied the evolution of the vegetable substance in a number of different
seeds, but seeks to identify therewith the uterus, the Fallopian tube, the
umbilicus, and the amnion, which naturally leads him to extravagant con-
clusions. Malpighi devoted special attention to the study of gall-formations
in a number of vegetable forms; he is fully convinced that they are produced
by insects, but on the other hand he found that the tubercles on the roots of
pulse plants are not produced by insects, though he failed to find any other
explanation of their origin.^ He also studied and speculated on a number of
other malformations in plants; with regard to the tubers in many plants,
he is of the firm opinion that they contain reserve nutriment. He is again
tempted, however, by the theory of the nutriment of plants, with which
he closes his work, to make dangerous comparisons with the conditions ob-
taining in the animal kingdom.
The other creator of plant anatomy
At the same time, however, as Malpighi submitted the first part of his vege-
table anatomy to the Royal Society, that society had sent to the printers
another work on the same subject compiled independently of Malpighi by
an English doctor, Nehemiah Grew. Born in i6i8. Grew was the son of a
clergyman who during the great Civil War joined the opponents of the Crown
and so, upon the return of Charles II, was deprived of his benefice. His son,
who was then an undergraduate at Cambridge, went (presumably for the
same reason) to continue his studies abroad. In 1671 he graduated at Leyden,
with a dissertation on the fluids of the nervous system. He then settled down
as a practitioner in a provincial town, but, thanks to the reputation he gained
by his work in vegetable anatomy, he was able within a short time to move
to London, where he applied himself to both medical practice and research,
eventually becoming secretary to the Royal Society. He died in lyiz.
As a scientist Grew concentrated almost exclusively upon vegetable anat-
omy, whereby his investigations at once acquire a different character from
those of Malpighi, in that the latter's constantly repeated comparisons with
human and animal anatomy are altogether lacking. Grew also studied the
^ It is only in our own time that it has been established that they are produced by
bacteria.
1 64 THEHISTORYOFBIOLOGY
construction of the fruit substance and the germination of the seed in a
number of plants, but in doing so he employs a terminology of his own,
nor does he borrow anything from animal anatomy; the word "paren-
chyma," which he invented, has been retained in vegetable anatomy. He
describes plants organ by organ; cells and vessels in the stem he discovered
independently of Malpighi and on the whole describes the anatomical details
more soberly and in greater detail, though with less fertility of ideas, than
the latter. He advanced a theory that the pistil in plants corresponds to the
female, and the stamen, with its pollen, to the male, and pointed out their
hermaphroditism, but, on the other hand, he entered into speculations upon
male and female ' ' juices ' ' in plants, which are of no interest nowadays except
from the point of view of mere curiosity. He voluntarily abandoned in favour
of Malpighi any claim to priority in regard to the discovery of the vascular
system in plants; on the other hand, Malpighi undertook a Latin translation
of Grew's writings. These two scientists improved vegetable anatomy so
far it was to take more than a century before any important addition could
be made to their work. Through them biology acquired its knowledge of
organized matter as being something peculiar in its structure; the idea of
tissue was established — for the time being, it is true, only in the sphere
of botany — and in the vegetable kingdom, also, the simple elements of the
tissues — the cells — were observed and described. It was, however, to be
nearly two centuries before the fundamental value of these achievements was
fully appreciated; true, both their contemporaries and the immediately suc-
ceeding age admired the exactness of their investigations, but it considered
the results more from the point of view of curiosity. All the greater admira-
tion is due to those scientists who at any rate guessed that here was to hand
information of the highest importance for the future of science. The fact
that their contemporaries failed to continue along the line they had laid
down was undoubtedly due mostly to the microscope's having at the same
time opened up a field in the sphere of animal anatomy of such considerable
scope and of greater immediate interest. The anatomy of the lower animals
in particular was an entirely unexploited field, possessing vast possibilities
for development, of which, indeed, splendid advantage was taken just about
that time.
Antony van Leeuwenhoek was born in 1631 at Delft in Holland and was
sent as a boy to Amsterdam to be trained for business. Having worked for
a time in the cloth trade, he returned to his native town and got an appoint-
ment with the municipal authorities,^ which must have taken up very little
of his time, since he was able to devote the greater part of his days to indulg-
ing his interest in the study of nature. As a research-worker Leeuwenhoek
^ His title was " Kamerbewarer der Kamer van Heerea Schepen van Delft."
SEVENTEENTH AND EIGHTEENTH CENTURIES 165
was a self-taught man; he had never received any scientific training, and as
he knew no Latin, in which language most natural philosophers of that
time generally published their works, he was unable in his old age to come
into contact with the scientific life around him; he had to depend entirely
upon himself. It was the remarkable phenomena revealed by the magnifying-
glass that fascinated him from the very beginning; he taught himself to grind
lenses, and by diligence and having a naturally delicate touch, he developed
this art further than any of his contemporaries. Sparing no pains to find out
new methods and combinations, he gave to his magnifying apparatus all
sorts of forms, some of them very strange; he tried glass, rock-crystal, and
even diamonds for his lenses; but the greatest advance he made was in the
manufacture of simple lenses with strong magnification; one such lens, which
is still in existence, is said to magnify as much as irjo times. As often with
self-taught men, he was extremely jealous of his inventions; he never sold
a magnifying-glass nor even lent one to anyone; on the other hand, scien-
tists who visited him were permitted to use a number of his instruments,
though never the most powerful. It is said that among his property there
were found more than four hundred microscopes and magnifying-glasses. A
number of them he had bequeathed to the Royal Society of London, of which
he was a member and which published most of his observations. Busily oc-
cupied to the last, Leeuwenhoek reached the age of over ninety; he died in
his native town in 17x3.
Leeuwenhoek' s investigations
Leeuwenhoek's collected writings have quite an extensive range, and their
contents are extraordinarily varied. The only connecting link that unites
them is the microscopical method; this Leeuwenhoek applied to literally
everything that came within his range of vision: crystals and minerals, plants
and animals. With respect to the last he developed no special microtomical
technique, but he studied and illustrated the details of what he observed.
This detailed study, however, he advanced further than anyone of his time,
and if he possessed the most powerful magnifying lenses known to his age,
he certainly had also the keenest eye. He took exact measurements of every-
thing that he examined; unfortunately there was in his time no unit of meas-
ure which could have served his purpose, so that he was compelled to select
such objects of comparison as he thought suitable — a hair, a grain of sand
— and to state his measurements in fractions, often thousandth parts, thereof.
He took careful notes of everything that he examined and sent them in the
form of letters to the Royal Society, to which he had been introduced by
his friend de Graaf, and of which he soon became a member. It often hap-
pened that one and the same letter contained a mass of different notes on
various observations he had made. It was undoubtedly to his great advantage
that he so seldom engaged in theoretical speculations, but only described
l66 THE HISTORY OF BIOLOGY
and illustrated what he saw; if at any time he starts theorizing, he generally
fails, but usually he appears conscious of his limitations and holds to the
realities which he knew so well how to master.
Biology has Leeuwenhoek to thank for a long series of facts of funda-
mental importance. His studies on the circulation of the blood deserve first
consideration. He has explained and completed the knowledge of the capil-
lary system which Malpighi originated, while he clearly proved that the
veins and arteries, each separately, are continued on immediately through the
capillaries and thus through them merge directly into one another. More-
over, Leeuwenhoek for the first time clearly recognized the blood corpuscles
and described them, first in the frog and then in man and a number of animal
forms. Malpighi had thought that he could distinguish in the blood "fat
globules," but he did not investigate the matter further, so that to Leeu-
wenhoek is due the honour of having really solved the problem. The same
is the case with the spermatozoa, which, it is true, a Dutch student by the
name of Hamm was the first to observe, but which Leeuwenhoek at any rate
studied more closely in a number of animal forms. In this connexion he made
thorough investigation into the fertilization of various animals, especially
fishes and frogs. In the frog he noted the spermatozoon's association with
the egg and believes — like Aristotle, as a matter of fact — that it is from
the male that the actual life comes; the female only provides, through the
egg, nourishment and powers of development. This he tries to prove by pair-
ing different-coloured rabbits : if a white female is paired with a grey male,
all the young will be grey like the father. Had he continued the experiment
through several generations, he would certainly have obtained other results.
— He has further observed a number of histological details of various kinds:
the stripes of the striated muscles, the structure of dental bone, the construc-
tion of the optic lens in man and the higher animals. No less remarkable are
his discoveries in the lower animal world. He thus discovered the Infusoria
and the Rotatoria in water; he explained the reproduction of ants, found
their true eggs, and showed that what had hitherto been called ants' eggs
are really the pupas of the insect. He very definitely opposes the hitherto
prevalent view that minute creatures of all kinds arise through putrefaction
or fermentation in inanimate matter. Instead he declares that even the small-
est animals possess reproductive powers and propagate solely by means of
them. In proof of this he demonstrated particularly the evolution of fleas
and aphids. If, finally, we add to this that he demonstrated the difference
between the structure of the stem of monocotyledons and dicotyledons in the
vegetable kingdom, this — by no means complete — sketch will have given
some idea of a life of activity which, without making any important theo-
retical contributions, advanced the knowledge of nature in an unusually high
degree.
SEVENTEENTH AND EIGHTEENTH CENTURIES 167
On the whole, Holland during the latter half of the seventeenth century
proved a centre of biological research. Of the many prominent scientists who
lived and worked in that country during that period it is possible to mention
only a few of the most important — those who in one way or another led
research into fresh directions.
Jan Swammerdam was born in Amsterdam in the year 1637. His father
was an apothecary who by saving had accumulated a considerable fortune
and was, moreover, interested in natural science. He possessed a natural col-
lection, which he augmented and looked after with great care. He had in-
tended his son to take orders, but as the study of nature seemed to be his
sole interest, he was permitted to study to become a doctor. After preliminary
studies in his native town he entered the University of Ley den in 1661. Even
then he had already proved a clever technician in the anatomical sphere,
and he rapidly acquired fame for his splendid dissecting and injection-work.
He formed a lifelong friendship with Steno, who was about the same age
and who happened to be visiting Leyden at the time; they worked together
and travelled together to Paris in order to continue their studies there. Here
Swammerdam found a new friend in the person of the King's librarian, The-
venot, a friend who throughout his life loyally assisted him in every possible
way. Returning to his native country, Swammerdam graduated at Leyden in
1667 with a dissertation on respiration and then settled down at his father's
place in Amsterdam. He had already earlier applied himself to the study of
the anatomy of the lower animals, and this interest now engaged all his
powers. During the short years that remained to him he achieved results
which not only left all his predecessors far behind, but actually remained
unexcelled for the space of more than a hundred years. In the mean while
his fortunes took an extremely unhappy turn. He contracted a malarial fever,
which, except for occasional intervals, never left him for the rest of his life.
At the same time the strain entailed on him by his work impaired his health.
Besides, he was of a passionate nature; his writings are full of bitter con-
troversy, and his quarrels about questions of scientific priority brought him
many enemies. On the other hand, he had several loyal friends, who stood
by him to the end. Worst of all, however, he fell out with his own family;
his father, who seems to have been an economical and surly old man, thought
that it was about time that his son, whom he had supported for more than
thirty years, applied himself to medical practice or some other profession
that might provide him with an income. In preparation for this, young
Swammerdam was sent into the country to recover his frail health, but he
spent night and day engrossed in his investigations, so that his health went
from bad to worse. After repeated quarrels his father finally deprived him
of all financial support. Swammerdam found himself in dire need and sought
in vain to sell his collections in order to buy his daily bread. Even his intel-
l68 THE HISTORY OF BIOLOGY
lectual powers had now waned; in about the year 1673 ^^ ceased to work
at his science and became abosrbed in religious contemplation. His old friend
Steno sought to take advantage of this state of affairs: at the price of the
same religious conversion which he himself had just undergone, he offered
Swammerdam splendid prospects in Florence. The latter refused, but instead
sought to cure his distress of soul by visiting Antoinette Bourignon, who
was very notorious at that time. She was an extremely gifted but hysterical
woman who in virtue of personal revelation desired to reform Christianity
on ascetic and mystical lines, and who, persecuted by both Catholic and
Protestant priests, wandered from country to country surrounded by a small
band of believers. Swammerdam joined this band, but was unable to find
the peace he sought; after leading a roving life for a couple of years he re-
turned in the deepest spiritual and bodily misery to his native country. There
at last he obtained, through his father's death, which occurred at the same
time, financial independence, but then quarrelled with a sister over the in-
heritance, which still further embittered his mind. In the year 1680 he found
repose in death, when not yet forty-three years old. In 1880 a beau-
tiful monument was raised over his grave and there was created to his
memory a fund, which is used for the purpose of giving prizes for research
work carried out in the spheres of learning in which he had studied.
Swammerdam's scientific activities thus lasted for only about six years,
during which period he published a few works of great value — particularly,
besides the dissertation above mentioned and an essay on the genital organs
of woman, a work on the anatomy of insects, in which he recorded his
earlier researches on that subject. His still unpublished manuscripts he be-
queathed to Thevenot; after the latter's death they passed through many
hands until they were finally purchased by the famous Boerhaave of Leyden
and were published by him together with the already printed work on insects
under the title of Bijbel der Natuure, in 1737. Although it thus came out more
than half a century after it had been written, this work was by no means
out of date; in fact, it was to be some time before its detailed anatomical
descriptions were improved upon — a proof of Swammerdam's incomparable
genius as an anatomist of invertebrate life. The title of the work was prob-
ably given to it by Boerhaave, but fully reflects the state of mind in which
the author found himself towards the close of his life. Nevertheless, religious
observations do not form any disturbing element in it; on the contrary, his
presentation is purely natural-scientific with the exception of a few con-
tributions of the religious moralizing character, particularly one that is a
reflection upon the short life of the day-fly. The undoubtedly valuable col-
lections on which Swammerdam based these studies were after his death
sold by auction and dispersed.
What still strikes the reader of Swammerdam's works is his mastery
SEVENTEENTH AND EIGHTEENTH CENTURIES 1^9
over even the most complicated details in the minute creatures he investi-
gated. This he could not possibly have acquired without a high standard
of knowledge of the technique of dissection, and, indeed, it was this knowl-
edge which excited the admiration of his contemporaries; visitors from far
and near were amazed at his fine instruments and the skill with which he
handled them — glass tubes drawn out to points as fine as hairs, by means
of which organs were spread out and canals injected; scalpules so fine that
they had to be ground under a magnifying-glass, and so on. Extraordinary
lightness of touch and unique powers of observation enabled him to utilize
the methods which he worked out, to which, finally, we must add his love
of research, for which he literally gave his life.
Anato7ny of insects
Swammerdam's great work in part contains a collection of anatomical mono-
graphs on insects and other invertebrate animals; particularly well known
is his exposition of the anatomy of the bee, which even Cuvier considered
to be unsurpassed, and further the head-louse, the day-fly, the rhinoceros-
beetle, the Helix pomatia, and many more. These monographs, however,
are all based on one theory of the evolution of insects and in connexion
therewith that of all living creatures. Supported in his investigations by
the development of a number of different insects' larv« to pupa and imago
and adopting a sharp controversial attitude towards Harvey, Swammerdam
declares that the insect does not undergo any transformation, but that merely
growth takes place of parts which already existed before. Again and again
this statement is emphasized, that no generation, but only an excrescence
of parts takes place, wherefore accident plays no part in the evolution of
the insect, but what takes place is predetermined. This evolutionary prin-
ciple is then applied to the development of the frog from the egg through
the various larval stages, and finally, though quite summarily, to the evolu-
tion of man, which is likewise made dependent on predetermined necessity.
Lastly, the evolution of the bud of plants to leaf and flower is compared
in detail with the metamorphosis of insects. In order to facilitate his analysis
insects are divided according to their metamorphosis into four groups:
(i) those that come from the tgg with all their feet complete — spiders,
lice; (i) the animal that has all its feet when it is hatched, but whose
wings develop later on — as, for instance, day-flies; (3) those from whose
egg comes a larva, either with or without feet, which becomes a pupa after
chrysalizing — as, for instance, ants, bees; (4) those in which a larva, like
the foregoing, becomes a pupa without chrysalizing — certain flies. In this
method of grouping Swammerdam laid the foundations of modern insect-
classification, which, as is well known, still rests to a great extent upon
the evolutionary history of insects. That he grouped spiders and even snails
and worms under his first category is not to be wondered at; all invertebrates
lyo THE HISTORY OF BIOLOGY
were at that time lumped together, and Swammerdam's point of departure
was from first to last not morphological, but evolutionary. But at any rate
he performed a service — as also did Leeuv/enhoek — in awakening interest
in these organisms, which had hitherto been regarded as existences not only
of a lower type, but also utterly incomparable with the higher; they still
arose, according to Harvey, by spontaneous generation, and this alone was
a proof that no conclusions could be drawn from them touching the life of
the higher animals.
Sivammerdam s pejormation theory
SwAMMERDAM showcd that on the contrary it was just the life-conditions
of the lower animals which, if viewed in a proper light, gave fresh stimulus
to the knowledge of life in its entirety. Particularly does he insist upon
this being realized in embryonic development. The theory which he ad-
vanced on this subject — growth of previously created parts instead of new
formation — came to exercise immense influence during the immediately
succeeding period: under the name of the theory of preformation or evolution
it entirely supplanted Harvey's theory of epigenesis. True, in its application
it was in its turn driven to sheer absurdities, particularly by certain scientists
who will be named later on, but when it first appeared, it was certainly
called for and marked an advance in biological science. In fact, it resulted
in the assertion for the first time of the obedience of ontogenetical evolution
to law; it definitely invalidated the old ideas of the spontaneous genesis
of lower animals; it established the fact that according to nature the off-
spring must resemble the parent, whereas in earlier times, practically speak-
ing, anything could arise out of anything — the legendary tales of women
who under the influence of witchcraft were delivered of kittens and puppies
instead of children had at any rate been discussed by certain scientists —
and, finally, it satisfied, as far as embryology was concerned, the contem-
porary demand for a mechanical explanation of nature. But it is true that a
century later the epigenetical theory was again to appear in a form that jus-
tified its acceptance — a change of which an account will be given further on.
There is a name that is worthy of mention by the side of Swammerdam
— that of his contemporary and friend Frederic Ruysch. He was born at
The Hague in 1638 of a highly respected family, his father being secretary
in the States General. Having when still young contracted an advantageous
marriage, he was in a position to apply himself at his own option to medical
research; he became a doctor at Leyden and a professor in Amsterdam, in
an appointment which he held for sixty-three years. Moreover, he had an
extensive medical practice. He died in 173 1 in his ninety-third year. His
long life, active to the last, is reminiscent of Leeuwenhoek's, as is also the
fact that his greatest service to the world consists in the employment of
technical methods, which, though he had not invented them, were neverthe-
SEVENTEENTH AND EIGHTEENTH CENTURIES 171
less improved by him. From his friend Swammerdam he had learnt the art
of using coloured wax for injections, and he acquired a masterly skill in this
method, such as few attained after him. He was able to fill out the finest
capillary vessels without either bursting or deforming them, and, besides, he
preserved the preparations thus carried out in a wonderfully natural manner.
And he was as jealous of his method as Leeuwenhoek was of his microscopes,
though really with far less excuse than the latter; the microscopes survived
the man who made them, while the method of injection went down with
its inventor to the grave. Even with regard to the value which his discoveries
had for science, the learned professor is no match for the untaught function-
ary, but Ruysch, in the application of his method, certainly did succeed in pro-
viding science with a mass of new facts, particularly in the sphere of human
anatomy. He discovered the bronchial arteries and the arachnoids of the
brain, besides which he studied and extended the knowledge of the iris and
retina of the eye; and, further, he compared the male and female skeleton and
investigated the changes made by age in the structure of bone. He made a
splendid collection of anatomical preparations, of which he published a
richly illustrated description. The objects were arranged in groups — human
organs, shells, minerals, and other things all together — in a manner which
in our time would be considered not only highly unscientific, but also utterly
lacking in taste.' His contemporaries, however, were ecstatic over it; for-
eigners visited the museum, and poets lauded it in verse. Tsar Peter of
Russia, who, as is well known, entertained almost childish admiration for all
products of technical skill, finally purchased the entire collection for thirty
thousand guilders, but naturally neither he nor any of his subjects could
make any use of it. A second collection, which was made later, was purchased
by the opponent of the Tsar, the Polish king Stanislaus Leszczynski. There is
now nothing left of these collections: it is only through Ruysch's books that
we in modern times can gain any idea of what he did. And it cannot be
denied that there was in him but little in the way of ideas, yet at the same
time extraordinary technical ability and quite a lot of humbug.
There was another Dutch physician of the same age as he, Reinier de
Graaf, who possessed far sounder qualities as a scientist. Born in 1641 of
Catholic parents, he studied at Leyden and at Angers in France, where he
took his doctor's degree. When still quite young he had won a great reputa-
tion, but owing to his faith he was prevented from obtaining a professorship
at Leyden, which was a strictly Protestant university, and he therefore
settled down as a practitioner in Delft. His unusually promising career was
cut short in 1673, when a serious illness deprived him of a happy domestic
life and the possibility of carrying on his intensive research-work. He had
' Thus there was amongst the groups the skeleton of a child holding a piece of injected
peritoneum like a handkerchief before its eyes.
lyi THE HISTORY OF BIOLOGY
already had time, however, to make important contributions to biology in
both the anatomical and the physiological sphere. His doctor's dissertation
deals with the pancreatic secretion. In it he shows how, by introducing a
canula into the duct of the pancreatic gland in a live dog, it is possible to
obtain some of the secretion for the purpose of closer examination — a
method which has since then been generally adopted in physiology.
The work on which de Graaf's fame principally rests, however, is his
study of the sexual organs, both male and female, but chiefly the latter. The
ovaries, of course, had already been described before, of both the higher and the
lower vertebrates; that they produced eggs in birds was known, but a great
many contradictory theories had been advanced on the subject of what kind
of function they possessed in man and the other mammals. The Aristoteleans
naturally supported their master's doctrine that the sexual product of the
woman is the menstrual blood and that otherwise the male semen is the
essential origin of the embryo, which the woman then nourishes and pro-
duces. De Graaf, on the other hand, after making a comparative study of the
ovaries of mammals and birds, came to the conclusion that the cell-like
protuberances already observed by Vesalius and Fallopio in the ovary of
mammals corresponded to the egg of the bird ovary, and that the process of
fertilization is similar in every animal type; just as a bird's fertilized egg in
the ducts of the ovary acquires albumen and shell, the egg of the mammal
becomes fertilized through the Fallopian tube, finds its way to the uterus,
and there develops further. The very word "ovary" was suggested by him;
hitherto the female sexual gland as well as the male had been called testis,
a word which he still employs alternatively with the new one. He definitely
rejects the assertion of the Aristoteleans that the embryo originates from the
man alone; in disproof of that assertion he cites many cases in which demon-
strably purely external characteristics have been inherited by the embryo
from the mother, both in human beings and in animals; even cases of extra-
uterine gestation are cited by him as proof that the embryo is derived from
the ovaries and not from outside. Likewise in regard to several other details
in the structure of the sexual organs he records valuable fresh observations.
These investigations of de Graaf's proved of fundamental importance,
although he was wrong in his assumption that the follicles in the ovary,
which now bear his name, correspond to the eggs in the ovary of a bird —
the true eggs of mammals were not discovered until a century and a half after
his death. Nevertheless, the explanation he gave of the actual phenomenon
of fertilization was of decisive significance for the future development of the
knowledge of this phenomenon. It was impossible, however, either for his
own or for the immediately succeeding age to reconcile his claim as to the
significance of the egg in embryonic development with the important part
that the spermatozoa should be assumed to play in the same process. And so
SEVENTEENTH AND EIGHTEENTH CENTURIES 173
we find developing, especially in the eighteenth century, the controversy
between ovists and animalculists, as the champions of the importance of the
egg and the spermatozoa respectively in fertilization were called — a con-
troversy which was to set science by the ears for decades. In some of the fol-
lowing chapters more light will be thrown on this controversy as well as on
the dispute between the respective champions of epigenesis and preformation
which was raging at the same time.
The close of the great period in the history of anatomy
With this, our account of the brilliant epoch in the history of biology repre-
sented by the seventeenth century comes to a close. It is perfectly natural that
towards the end of that century, and during the decades immediately follow-
ing, interest should wane in just those spheres in which progress had been
greatest; the forced march had to be followed by a period of mustering of
forces and reflection, during which the results achieved had to be weighed
from the theoretical point of view and classified. It is therefore worth while
considering what were the solutions which the next age sought to give to
the theoretical questions that had arisen in connexion with the great practi-
cal advance made in the field of anatomy and experimental biology. In the
following chapter some samples will be given of theories of this kind.
CHAPTER V
BIOLOGICAL SPECULATIONS AND CONTROVERSIAL QUESTIONS AT
THE BEGINNING OF THE EIGHTEENTH CENTURY
AS HAS BEEN POINTED OUT it! the fofcgoing, the power of the authori-
ties of antiquity was broken during the seventeenth century as the
L result of a series of brilliant scientific discoveries; in its stead nat-
ural scientists based their researches upon the knowledge of the mechani-
cal subjection to law which prevails in nature. However, the need was
felt for a definite and uniform conception of nature such as Aristoteleanism
undeniably possessed and which was lacking in the new systems of thought
which took its place. In actual fact these systems, whether they emanated
from Descartes, Spinoza, or Leibniz, were quite as dogmatic as Aristote-
leanism; they were pure thought-structures, which, although based on the
new natural science, were yet by no means capable of satisfactorily solv-
ing the problems to which that science gave rise. As far as biology is con-
cerned, while it is true that physicists like Borelli or Perrault had been
able with the aid of the newly-discovered mechanical laws to find solutions
to a number of pure problems of motion, yet as soon as more complicated
processes in the organism, such as the digestion, the circulation, or sense-
impressions, had to be considered, the mechanical principle was found want-
ing; nor had the other branches of physics and even chemistry as yet reached
such a state of development that they could be employed as a means of ex-
plaining such phenomena as those just mentioned. In these circumstances
many a scientist was content merely to study the new facts which had been
brought to light as a result of improved experimental technique, but there
were others who devoted their lives to seeking firm ground on which to base
a uniform explanation of life-phenomena. In modern times it is not easy to
appreciate the difficulties with which these biological thinkers had to con-
tend in their efforts to reconcile the individual results of past research work
under one common point of view. Uniformity in the conception of nature
in our day, of course, rests essentially upon the law of the indestructibility
of energy, to which may be added, in the field of biology, the doctrine of the
cell as a unit of life. But the theoretical natural science of the seventeenth
century tended, instead of to these ideas, to the assumption of the existence
of an unknown force as the origin of life and a basis for its continuance. This
force could then be conceived of as something either purely mechanical or
174
SEVENTEENTH AND EIGHTEENTH CENTURIES 175
more idealistic; in the former case one was bewildered, since mechanics can-
not provide the answer to more than a small fraction of the questions which
the new discovery brought to light; while in the latter case there was the
risk of reverting to mysticism in one form or another. These natural-scientific
speculations from the beginning of the eighteenth century, which we shall
now discuss, originated, curiously enough, less from the anatomists and
biologists than from the medical practitioners, who sought to base their
medical treatment on a general theory of the functions of the body. Of these
latter scientists some few have exercised a radical influence even on the gen-
eral development of biology and therefore deserve to be mentioned in this
connexion.
Thomas Sydenham lived, it is true, entirely in the seventeenth century —
he was born in 162.4 ^^'^ ^^^'^ ^^ ^^^9 — ^^^ ^^^ influence did not really make
itself felt until after his death, and, indeed, it has increased still more since
then. He belonged to a good country-family and studied for a time at Oxford,
but upon the outbreak of the Civil War he joined the Parliamentary party
and became an officer in its army. Afterwards, however, he continued his
medical studies, took a low medical degree, and settled down in London as
a pracititoner; he did not obtain his doctor's degree until he was over fifty
years old. Personally Sydenham enjoyed a great reputation; he counted among
his friends such people as the chemist Boyle and the philosopher Locke. On
the other hand, opinions differed as to his capacity as a physician; his auda-
cious ideas required time before they could penetrate the ordinary mind.
Nowadays he is universally regarded as one of the pioneers of medical
science.
Sydenham s medical doctrine
In the seventeenth century London was a very unhealthful city; one plague
followed another in rapid succession. It was these epidemics that inspired
Sydenham to work out his medical theories; he studied the symptoms of
the various diseases and endeavoured by that means to characterize the dis-
ease itself in the same way as the botanist describes a plant-species. "That
botanist would have but little conscience who contented himself with the
general description of a thistle and overlooked the special and peculiar char-
acteristics in each species." He places a higher value on this exact study of
nature than on any theories; this study should take into account all factors
affecting the disease in its entirety. Even the season of the year when the dis-
ease is most widely dispersed should be carefully observed, as indeed all other
conditions that may influence the plague as a whole. On the other hand,
purely individual variations in particular cases are of minor interest. Like
Hippocrates he considers that it is the nature of the patient that cures dis-
ease; it is therefore not so much worth while worrying about trying to
diagnose the disorders in the fluids of the body on each occasion as to try to
1/6 THE HISTORY OF BIOLOGY
discover a treatment that may assist the working of nature. What he really
means when he talks of "nature" is not at all clear — whether it is a com-
bination of the individual's life-manifestations or some special life-force;
similarly, one does not gain a very clear explanation of the ideas he borrowed
from Hippocrates concerning the fluids of the body and the balance or dis-
turbances therein. His general conception of nature is on the whole purely
empirical — in this he was influenced by Bacon, whom, indeed, he quotes
with admiration. In certain cases, it is true, he can form quite daring hy-
potheses, but as a rule he consistently applies his principle as to observa-
tion's being the only source of knowledge in disease.^ This principle has
indeed been adopted by posterity, but he also exercised a powerful influence
on the medical and biological thinkers of the immediately succeeding age,
although these latter could not restrain themselves within the limits which
he laid down for research, but went further afield in the world of
hypothesis.
Among these medical researchers who formed general theories of impor-
tance to the development of biology, two men are conspicuous at the begin-
ning of the eighteenth century who, born in the same year and working in
the same town, yet proved in all essentials strangely contrasted. These two
were Hoffmann and Stahl.
Friedrich Hoffmann was born in Halle in 1660, the son of a wealthy
physician. At the age of fifteen he had the misfortune to lose both his parents,
who died of the plague, as well as his inheritance, as a result of a fire, and
thus early had to fend for himself. He was given an opportunity, however,
of studying medicine at the University of Jena, where a highly reputed
representative of the chemical and medical research of the period, G. W.
Wedel, was his teacher. Having further studied at Erfurt, he took his de-
gree at Jena, spent some time in England, where he made the acquaintance
of Boyle, and then set up as a practitioner in a couple of small German states
until, in 1693, he was called to an appointment at the newly founded uni-
versity in his native town of Halle. There he spent the rest of his life as a
professor, with the exception of a couple of years which he spent at the court
in Berlin. His work both as a teacher and as a physician was crowned with
extraordinary success: among his numerous pupils were included even old
doctors who sought to complete their training with him; is a practitioner
he was resorted to by high and low and was overwhelmed with consulta-
tions and loaded with brilliant honours. Considerate even towards those of
different opinion, of an affectionate, sympathetic nature, he was himself
universally beloved. He died in 1741, in harness to the last.
^ In his own circle his contempt for theories seems sometimes to have been expressed in
a somewhat original way ; thus, a colleague who asked for advice regarding the choice of medical
literature was told to read Don Quixete.
SEVENTEENTH AND EIGHTEENTH CENTURIES 1 77
Hoffmann's practical ivork and bis theory
Undoubtedly Hoffmann's services to science lie principally in the sphere of
practical medicine. He described several diseases hitherto unaccounted for;
both in theory and in practice he insisted upon accurate diagnosis based upon
natural-scientific principles, considerate treatment of the sick, and simple
medicines. He himself made up and sold at a great profit quite a number of
preparations, which still play their part in popular medicine. He was a very
productive writer on medical subjects in every conceivable specialized sphere,
but he also tried to combine in one general theory of the functions of the body
the principles at which he had arrived in the course of his work, and this
theory has not been without its importance for the general development of
biology. It takes as its starting-point the so-called chemiatric theories preva-
lent in the seventeenth century, which ultimately originated in Paracelsus's
fantastic speculations as to the human body's being composed of quicksilver,
sulphur, and salt, and in conformity with its original sought to explain the
functions of the body as essentially phenomena of chemical change, for which
purpose recourse was had either to the mechanical theories of the movements
of the body, described in the foregoing, or else to the mass of mystical specu-
lations still available at that time, to fill up the gaps in the proposed system.
Hoffmann takes his stand at the very start on chemico-mechanical ground. He
himself was a clever chemist and besides possessed a complete mastery of the
anatomical literature of his age, in which sphere both Borelli and Perrault
had some influence on him; and, finally, he had not neglected the discoveries
of either Newton or Leibniz. He began with the principle that matter and
motion form the foundation of existence; the body is a machine, which is
kept going by the circulation of the blood. Life is thus a purely mechanical
process, from whose functions the activities of the soul can be excluded;
when the body dies, it is not the soul that leaves the body, but the body that
abandons the soul, so that the latter can no longer use the organs of the body
as its tools. The movement of the blood is caused by the heart; the latter's
action, again, is regulated by the movements in the nervous system, in the
fibres of which there circulates a fluid, "sfiritus animalis," which is formed
of extremely light ether-particles and is produced in the brain and by its
movements induces and regulates the muscular functions, sense-impressions,
and alimental processes. The power of the blood to maintain life is due to
the fact that it contains a "spiritus" formed of the ether constituents of the
air and the sulphurous element in the blood. Chemically, in fact, the com-
ponents of the blood are partly sulphurous, partly ethereal, and partly
earthy; the sulphur element is the cause of the warmth of the body, both the
natural warmth and that increased by inflammation, which is induced by
the sulphur particles easily becoming extremely mobile through the action
of the ether. The function of the lungs is to mingle the component parts of
lyS THE HISTORY OF BIOLOGY
the blood, besides which the inhaled air conveys to it fresh ether particles,
which augment its power to keep the mechanism of the body working. It
would take too long to record here the complicated accounts of the produc-
tion and dispersion of the nervous fluid; it may just be mentioned that the
finest and most vitally essential part of the fluid is said to emanate from the
cortex of the great brain. The male semen is closely akin to the nervous fluid,
and its function is thus to give life to the egg, so that it may start developing.
Hoffmann, having thus described the mechanism of the body, declares
that man naturally possesses an immortal soul, given him by God; the will of
this soul controls the movement of the body, and through it we understand,
think, and act. Following many other old authors and supported by the Holy
Scriptures, he divides man into three "principia" — namely, corpus, spiritus,
and anima — that is, the body, the above-described nervous fluid, and the con-
sciousness. But besides these man possesses a higher "substance," which the
ancient philosophers called mens and which Scripture names the image of
the spirit of God; this substance makes use of the consciousness's impression
of things and forms them into ideas; false sense-impressions may be rectified
by clear reason, but a mass of confused sense-impressions causes madness.
Concussion of mind may also disturb the circulation of the blood and produce
a condition of sickness in the body. But Hoffmann resolutely denies that
the movement and function of the body originate in the soul; "although the
human soul possesses a certain limited influence over the bodily parts, never-
theless medicine both in theory and in practice is pure mechanics, in that it
is based upon purely mechanical principles — namely, motion and matter."
The inconsistencies and the arbitrary constructions of thought in this attempt
to form a mechanical conception of life-phenomena will be easily realized by
the modern reader, but should not in any way detract from the respect due
to this attempt — which at any rate is based on very substantial experiences,
considering the age — to find a natural connexion in the life-process. The
assumption of an immortal soul is explained by the fact that Hoffmann was
a devout Christian with a markedly pietistic temperament; Halle was the
source and centre of pietism, and Hoffmann was a warm friend of its founders,
Spener and Franke. He also displays in many places a naive childlike piety,
as when he dedicates one of his books to "The Holy Trinity, the Supreme
Physician." Thus we have here a proof that a mechanical conception of life
and ancient theological dogmas were formerly capable of being reconciled,
which would hardly be considered possible in our own day.
One of Hoffmann's first steps when elected professor at Halle was to
bring about the appointment of an old fellow student from Jena, Georg
Ernst Stahl, to be assistant professor of medicine; Hoffmann retained for
himself the position of teacher in practical medicine, while Stahl took over
the theoretical side. Stahl was born in 1660 of a Protestant family at Ansbach
SEVENTEENTH AND EIGHTEENTH CENTURIES 179
in Bavaria, receiving a strictly religious upbringing, which left its mark on
his entire life. He studied at Jena, where be became a doctor and for a time
gave lectures. After having been for some years court physician at Weimar,
he came, as mentioned above, to Halle and taught there for about twenty
years. At first his relations with Hoffmann were in every way friendly, but
gradually the good feeling between them changed, and, finding that Hoff-
mann's personal superiority excluded all possibility of competition, Stahl
resigned from his professorship and in 1716 accepted an appointment as
physician to the court in Berlin. He died there in 1734. Hoffmann and Stahl
possessed their pietistic devoutness in common, but otherwise they were
highly contrasted: Hoffmann, of stately build, lovable, and popular; Stahl,
in his appearance insignificant, in his manner austere and inaccessible, in-
tolerant towards his opponents, and bitter in controversy. At any rate he
was a sincere seeker after truth, who was honest enough — a quality other-
wise not very common amongst scientists — when he changed his opinion,
openly to admit the incorrectness of his former views, and he likewise
possessed that rare habit of gratefully acknowledging his predecessors'
contributions to the problems he dealt with,
Reformer of chemistry
As a scientific writer Stahl was, like Hoffmann, extraordinarily productive
and he dealt with a considerable number of different medical problems. As
a scientist he was undeniably superior to his rival; in fact, his name is one
of the foremost in the history of the natural sciences — principally on ac-
count of his work as a chemist. At the close of the seventeenth century there
was still being commonly taught at the German universities the subject of
alchemy — belief in the transformation of metals, in the philosophers'
stone, and all the rest of the mediasval mysticism which in western Europe,
thanks to Boyle and his successors, had already been disestablished. Even
Stahl had begun as an alchemist, and in his earliest writings he discusses
the usual alchemistic problems, but by his own efforts he undeceived himself
and thereafter never hesitated to point to the treatises of his youth as a
warning. That uniform conception of the changes in nature which the
alchemists sought to produce by means of their mystical speculations Stahl
now endeavoured to attain by a comparison of those processes that take
place in combustion on the one hand and the calcination of metals on the
other. Finally he obtained a common ground of explanation for these phe-
nomena by postulating the existence of a fluid substance, phlogiston, in both
combustible substances and metals; in combustion phlogiston disappeared
from the burnt material, as it did also from the metal in calcination; the
metal calces were thus like the metal, minus phlogiston. If the metal calces
were heated with a substance containing phlogiston — as, for instance,
coal — the metal was recovered by the reintroduction of phlogiston. —
l8o THE HISTORY OF BIOLOGY
This theory rendered possible a uniform conception of a number of processes
of conversion in inorganic nature; it constituted a working hypothesis
which had a great influence upon the science of chemistry in succeeding ages
and made the eighteenth century a period of brilliant achievement in chemi-
cal history; names such as Priestley, Bergman, Scheele, bear witness to the
progress made in chemistry when the phlogiston theory was dominant; and
when eventually Lavoisier, by introducing the weighing method, proved
that the theory was untenable and substituted the idea of oxidation for cal-
cination, the new theory could be applied directly to the discovery that had
been made when the old theory prevailed. Were it only for the advance he
thus brought about in chemistry alone, Stahl would deserve a place in the
history of biology, which has been so essentially dependent upon the prog-
ress of chemistry, and indeed will always be so.
StahVs medical theory
What constitutes Stahl's principal claim to be mentioned as a biologist,
however, is the theory of life which he expounds in his great work Tbeoria
rnedica vera, in which he seeks to formulate a general theory of the human
body and its functions, both in its normal state and in sickness. He himself
has declared, and it has been repeated after him, that his chemical theories
exercised no influence upon his ideas on the subject. This is true in so far as
he does not — like his predecessors amongst medical chemists, Paracelsus,
van Helmont, and others — base his entire conception of the human body
upon speculation as to its chemical composition, but, on the other hand, the
essential part of his work gives ample proof that chemistry is the science on
which he bases his ideas. Above all, he is no anatomist; he scorns the result
of ordinary macroscopical anatomy and he can hardly find words to express
his contempt for Leeuwenhoek's and de Graaf's microscopical investigation
of the sexual products; he likewise strongly contemns the discovery of the
capillary system, the existence of which he simply denies. On the other hand,
he displays a very keen interest in the "mixing {nnxtio)" of the body and its
parts — that is, their chemical composition — and he believes that a true
conception of the phenomena of life should be based on the knowledge of
this ' 'mixtio.' ' Indeed, it is in this direction that he has performed his greatest
services to biology.
The first chapter of Stahl's principal work, mentioned above, is entitled
"An Examination of the Difi^erence between Mechanism and Organism."
This title might well hold good for the whole of Stahl's literary work
on general science; the contrast mechanism--organism is to him the main
point in both biology and medical science; he discusses it from every con-
ceivable point of view, and in support of his views thereon cites a number
of arguments, both good and bad. The main argument, which he repeats
again and again in proof of his theory, is that organism is something funda-
SEVENTEENTH AND EIGHTEENTH CENTURIES l8l
mentally different from mechanism, that consequently the mechanical
physiology which his contemporaries universally accepted must be utterly
repudiated. In the living organism the soul is the essential part; the body
exists for the sake of the soul and is controlled by the soul. As a proof of
this assertion he quotes, to start with, a number of ancient Aristotelean
arguments on the finality of the structure of the body. Further, he declares
that the existence of the body is due to a thing which is in itself foreign to the
essence of the body, but, on the other hand, is akin to the essence of the soul,
owing to its immateriality — namely, motion. The soul's function consists
in going from object to object and comparing them, and the maintenance
of the body by means of mental activity and constant moving goes on, sub-
ject to the will of the soul, as the result of motions suited to the objects that
the soul requires. The fact that Stahl thus calls motion "thing" and com-
pares it with the soul in contrast to the body proves that at any rate he had
learnt nothing from Galileo and Newton. If, then, we find in this and other
similar arguments the utter hollowness of Stahl's philosophical speculations,
he has on other occasions an exceptionally keen eye, trained through his
chemical studies for the essential in the composition of organism. As some-
thing essential to all the constituents of the body he points out the extreme
easiness and rapidity with which they are chemically decomposed. This
property evidently made a great impression on him; he constantly reverts
to it and searches for an explanation for it, but it is obvious that, with the
fundamental ideas that he once embraced, it is always the soul which ulti-
mately keeps the body together and prevents it from disintegrating. This
easy deccmposability is considered to be due to a very complicated chemical
combination in its constituent parts — a fact that differentiates it from
ordinary chemical associations. The chemical quality is different in different
forms of life and peculiar to each individual. Finally, the constituent parts
of the body possess, besides their chemical quality, a special "texture" and
"structure": the former an arrangement of the smallest parts of the body,
the latter a combination of the elements thus formed, these two factors being
characteristic for every living being. "Living body is nothing else than that
which has structure," he declares. It is hardly necessary to lay special stress
on the fact that as a result of all this investigation into the chemical nature
of organism Stahl advanced science a long way; both the complex composi-
tion and the resultant easy decomposability of the constituent parts of the
living body are indeed facts of fundamental importance for modern biology,
and of still greater importance is his postulate that structure is something
peculiar to the living organism in contrast to dead natural objects. Here
Stahl has without doubt had some presentiment as to the significance of
tissue structure as a basis of life in all its forms; that he was unable to follow
up the idea to a conclusion of immense value to science was certainly due
iSz THE HISTORY OF BIOLOGY
to his lack of interest in anatomy. Nor, indeed, did his contemporary age
realize the importance of this question; it was not until sixty years after
Stahl's death that Bichat, basing his results on anatomical studies along
many different lines of inquiry, established the vital part played by the
tissues in maintaining the functions of the body, but, as we shall see later
on, he had come from a school in France that adopted and developed Stahl's
ideas.
Doctrine of the soul as cause of life-phenomena
The theory of Stahl's which aroused most interest in his time — that is,
which evoked most applause and most controversy — was his doctrine of
the soul as the cause of all life-phenomena, as their one supreme condition
and their final aim. This "animistic" conception of the structure and func-
tions of the body, according to which every manifestation of life, whether
it is a question of the absorption of food, the blood-circulation, the processes
of secretion and excretion, or simple movements from one place to another,
muscular activity and sensations, takes place exclusively for the sake of the
soul, is induced by it, controlled by it, and pursues its normal course thanks
to it — this theory, so utterly opposed to the contemporary mechanical con-
ception of life, was in reality the one main factor for Stahl, the very founda-
tion on which he built up his medical system. For Stahl aimed at creating
a new medical science, and his speculations in common biology were in-
tended merely to lay the foundations of that science. Naturally, the dis-
eases of the body are also caused by the soul; if it relaxes its control of the
body or any part thereof, there at once ensues general or local decomposition
of the inconstant chemical associations of which the body is made up, and
sickness or death results. And Stahl does not hesitate to follow up this theory
to its ultimate conclusions: if the soul desires to do so, it can naturally keep
the body whole, but, as it happens, the soul is wayward, inconstant, and
inconsiderate, and the body has to suffer for it. The soul of animals possesses
in this respect less freedom of action than the human soul, with the result
that animals are less often sick. One would suppose that in these circum-
stances any kind of medical treatment would be superfluous, since it has to
deal with the body, which is in any case powerless, but Stahl does not draw
this conclusion; like the homoeopaths of a later period, however, he pre-
scribes remedies having a mild action, with which he believes it possible to
help the soul in its functions to the improvement of the body; violent
remedies, such as quinine and opium, he deprecates. There is one con-
clusion that he draws from his system which does him honour — namely,
when he prescribes mild treatment in mental cases; otherwise the physicians
of his age, even the most humane, generally employed violent and sometimes
brutal methods in their attempt to drive out the mental disease from the
unfortunates who were thus afflicted.
SEVENTEENTH AND EIGHTEENTH CENTURIES 183
Those of Stahl's contemporaries who adopted his ideas were at any rate
not compelled to associate themselves with the peculiar theories referred to
above. His criticism of that age's mechanistic conception of life is indeed
often of such penetrating keenness that it must have proved attractive to
those who sought to probe the contemporary controversial problems in that
sphere. Especially does he inveigh against the theories of these "vital
spirits" on which his opponents' explanations of the phenomena of life
rested and which they could not possibly do without. Compared with these
theories his soul-theory was at least simple and easy to comprehend; it cleared
up satisfactorily enough the question of the relation of the psychical phe-
nomena to the material, a problem on which all previous attempts to explain
mechanically the phenomena of life came to grief. Stahl also had a sharp
eye for other weaknesses in the contemporary explanations of life and demon-
strated their inanity, as, for instance, the pan-sperma theories that were so
common at the time. Besides his above-mentioned keen analysis of the con-
trasts between living and inorganic natural objects, which is only briefly
summarized here, these critical contributions relating to the controversial
biological questions of his age constitute Stahl's great service to science.
This is, it is true, counterbalanced by his vague, yet subtle, natural philosophy,
which has also been but briefly recounted here, and the understanding of
which is rendered all the more difficult by a very obscure and badly arranged
method of presentation. He gained many followers among his contempora-
ries; several of his own pupils gave practical demonstrations of the dangers of
regarding the soul as an instrumental component in the functions of the body
and the treatment of disease by indulging in extravagant speculations along
mystical and theosophical lines. The valuable parts of his theories were
most strictly adhered to and most faithfully developed at the University of
Montpellier, where an entire school of physicians embraced his ideas. Among
his opponents may be especially mentioned, besides his old friend Hoffmann,
Leibniz, who in a contentious pamphlet sharply inveighed against his con-
tempt of anatomy, chemistry, and other exact methods of research, and, from
the standpoint of his own monad theory, rejected Stahl's theories of the soul
and motion as being separate from the material part of living beings and as
factors operating independently thereof. The influence that Stahl had on the
development of biology in later times may at first glance seem small; in-
directly, however, he has certainly been of greater significance than many of
those who are more frequently quoted. Among those who have openly
acknowledged their indebtedness to him may be mentioned such a compara-
tively well-known scientist as the embryologist Caspar Friedrich Wolff.
The man who, of the medical and biological theorists of that time,
undoubtedly enjoyed the highest reputation among his contemporaries was,
however, Hermann Boerhaave. He was born in 1668, the son of a country
184 THE HISTORY OF BIOLOGY
parson, near Leyden in Holland; and there he studied, first of all, theology;
but after becoming acquainted with Spinoza's theories he soon put an end
to all idea of entering the clergy. So he had to look about him for a new-
means of livelihood. After taking a degree in philosophy at Leyden he moved
to the small university of Harderwijk and there very quickly passed a medi-
cal examination, after which he settled down in Leyden as a practitioner
and teacher. At first he had a hard struggle, but he assiduously carried on his
profession, and his reputation rose year by year until he was finally elected
cO the first chair of medicine at Leyden and became universally acknowledged
as the foremost physician in Europe. In that position he acquired an influ-
ence such as few have ever possessed before or since; his advice was sought
not only from every corner of this hemisphere but even from the most distant
parts of the East. He made a vast income and died a multi-millionaire. These
successes were made possible owing to his brilliant gifts and, in spite of
lifelong physical ill-health, his unfailing energy. But above even these
merits his contemporaries valued his noble character; he lived extremely
simply, while he used his great wealth to render help to the poor and sick
and to give generous support to science; thus, he rescued Swammerdam's
writings from destruction and enabled Linnaeus to carry out his work in
Holland; he was friendly and modest in society, but when the necessity
arose, he could stand upon his dignity against even the highest in the com-
munity. He died in 1738, having during the last years of his life had to give
up his professorial duties owing to ill health.
Boerhaave' s theory: limitation of natural-scientific research
Boerhaave's attitude towards the general biological problems of his time
was undoubtedly dictated by the fact that he had studied Spinoza in his
youth and was throughout his life a keen admirer of Sydenham. Reminiscent
of the former is his clearly and vigorously expressed characterization of the
relation between body and soul. In man everything that involves thought is
to be ascribed entirely to the soul as its starting-point. Whatever, on the
other hand, involves extension, impenetrability, form, or motion, must be
referred entirely to the body and its motion. Again, he is reminiscent of
Sydenham in his realization of the limitations of natural-scientific research.
"The investigation of the ultimate metaphysical and the primary physical
causes is neither necessary nor useful nor possible for a physician. Examples
of these causes are: the elements, the first forms, the origin of procreation,
movement, etc." This quotation, moreover, shows his decidedly practical
nature, also resembling Sydenham's. He actually placed it as the foremost
aim of his science to create capable practical physicians. To gain this end,
however, he considered that a grounding in general science was indispensa-
ble, and he therefore made a close study of the general structure and func-
tions of the body, his work being based on what was a rare thing in those
SEVENTEENTH AND EIGHTEENTH CENTURIES 185
days, a thorough knowledge of the entire range of the then known medical
and biological literature. The common biological, or, as he termed it, phys-
iological, section of his principal work, Institutiones medicce, gives the im-
pression, owing to the mass of second-hand information that he imparts
when quoting his sources, of being to a certain extent a compilation — in
fact, it was published with the expressed intention of providing a handbook
for instructional purposes — but the ideas he presents in it are at any rate
thought out on entirely original lines, and the whole work seems, in com-
parison with Hoffmann's or Stahl's theoretical speculations, strikingly
modern. The abstract theories are, as a matter of fact, entirely thrust into
the background in favour of a close analysis of all the known facts relating
to the functions of the body. First he describes the digestion, starting with a
detailed account of mastication; then the functions of the digestive canal
and its glands; then the circulation of the blood, and respiration, the brain
and nervous system, several glandular systems, the muscles, the skin, sen-
sations, and reproduction. From a purely anatomical point of view the
presentation does not on the whole differ from the results achieved in modern
times; we find here that he has taken full advantage of every step of progress
made by such people as Borelli, Malpighi, and Ruysch. In particular
Ruysch's careful dissections and injections Boerhaave, who, indeed, was a
personal friend of his, was able to take advantage of in a masterly way.
Hij" mechanical conception of life
When he comes to explain the functions of the different organs, he bases his
ideas on a strictly mechanical conception: the action of the body is motion;
"the power to exert movement is called function, which takes place in ac-
cordance with mechanical laws and only by them can be explained." Thus
both the disintegration and assimilation of food in the body are purely
mechanical — ■ he denies that the gastric juices have any chemical reaction —
the principal agent is the body's own heat and the constant movements of
the digestive canal and its surrounding organs, but the nervous fluid also
plays a predominant part in the functions of the body. With regard to the
question of the "cooking" of food in the digestive canal, as assumed by
ancient authors, Boerhaave takes up a somewhat sceptical attitude. On the
other hand, he believes that acrid and unsuitable food-substances become ex-
cluded by contracting the openings of the chyle vessels into the bowel. Such
food as has been taken into the chyle vessels is conveyed through the thorax
to the venous system; there blood and chyle are mingled, and this mixture
becomes complete through the blood's passing into the lungs, whose porous
structure serves to render the mixture as thorough as possible. In a conten-
tious article written against Borelli, who believed it to be the case, he denies
that the air from the lungs passes into the blood; Boerhaave is unable to
explain why it is that living creatures cannot breathe in an unventilated
l86 THE HISTORY OF BIOLOGY
room. The brain also serves to purify the blood, which passes through it for
that purpose. Moreover, the cerebral cortex collects from the blood its finest
constituents, which give rise to the fluid that is conveyed from the brain
through the tubular nerve-threads out into all the different parts of the body
and induces movements in them. In particular Boerhaave inquired deeply
into the problem of muscular contraction and its relation to impulses derived
from the nervous system; he gives an account of an experiment to show that
muscular action is dependent upon the nerve and he considers that this in-
fluence of the nerve is due to the flowing of fluid from the brain. With regard
to the mechanical action of the muscles, Boerhaave highly commends Borelli's
mechanical investigations; the affluxion of the nervous fluid he believes takes
place in accordance with Mariotte's law.^ Boerhaave gives a detailed de-
scription of the structure and function of the genital organs, which is based
on the discoveries of Leeuwenhoek and de Graaf. He holds that the sperm
is "refined" blood; its small, living "animalcula" contain rudiments of the
organs of the future embryo; as eggs he regards the follicles in the ovary,
in this following de Graaf; conception takes place as a result of the "living
elements" of the sperm penetrating the pores of the egg.
As a whole Boerhaave 's biological theory must be considered to come
far nearer our modern ideas than either Hoffmann's or Stahl's — this both
on account of what he knows and above all on account of what he considers
it impossible to know. His insight into the limitations of natural science
really testifies more than anything else to his greatness; as regards facts, we
cannot expect of him more than it was possible for his age to attain. But it
is just his deliberateness that it has been difficult both for his contemporaries
and for posterity to understand; the desire to solve the ultimate riddle of
life has again driven the philosopher beyond the limits of what science can
attain with the means available. We shall leave Boerhaave, clear-sighted
and conscious of his own limitations, and shall proceed to consider a scientist
who sought to solve the riddle of life along speculative lines and who ex-
pended on this endeavour one of the richest and most fertile geniuses known
to history — namely, Swedenborg.
The son of Bishop Jesper Swedberg, a famous hymn-writer and preacher
of the Swedish Church, Emanuel Swedenborg was born in 1688 and received
a thorough school and university education at Upsala, where he grew up;
having completed which, he spent several years in England and on the
Continent, studying principally natural sciences, both theoretical and ap-
plied. Having returned home, he served as a military engineer during the
last fighting years of Charles XII, then became assessor of the board of mines,
^ Boerhaave undoubtedly refers to the hydrostatic experiment which goes by the name of
Mariotte's bottle; how its phenomena are to be applied to the nervous and muscular functions
is, however, not clearly stated.
SEVENTEENTH AND EIGHTEENTH CENTURIES 187
and was elevated to the nobility,' displaying during the next decades inde-
fatigable energy as an official, member of the House of Nobles, and scientific
writer. Then during the years 1744-5 ^^ underwent a severe spiritual crisis;
after repeated phases of alternate depression and exaltation he beheld in a
vision the Saviour Himself and learnt from Him that he was henceforth to
devote himself entirely to spiritual matters. He at once resigned his post of
assessor and devoted his whole life to spreading the new doctrine that he
believed he had received direct from heaven through repeated spiritual
revelations. Pestered by the priesthood in his native country, he lived his
last years mostly abroad, and died in deep poverty in London in the year
1772., misunderstood by his own age, but honoured as a religious founder by
a small group of believers.
Swedenborg's natural-scientific works are extraordinarily extensive; he
published books on mathematics, physics and chemistry, geology and cos-
mology, anatomy and physiology, and, besides this, much of what he wrote
remained unprinted and has not been published until our own time, as, for in-
stance, his anatomical work De Cerebro, which, contains his most important in-
vestigations regarding the brain. All these works are full of ideas and genius,
the true value of which was not appreciated until our own day, but which,
on the other hand, contain very little in the way of original observations.
He himself considered that he possessed more talent for thinking about
already existing facts and their interrelation than for making observations
of his own; but for the very reason that he did not support his speculations
upon facts which he himself had observed, he ran the risk of letting his
thinking be influenced by that attraction for the mystical which he had
always felt and which had been encouraged by the religious environment of
his childhood. Among the students of nature who thus impressed him must
especially be mentioned Olof Rudbeck, who in Swedenborg's youth was the
predominant figure in the University of Upsala and from whom he learnt
not only his love of nature, but also a tendency to many-sided activities and
fantastic conclusions. As we have already seen, Rudbeck was an upholder of
the seventeenth century's mechanical conception of natural phenomena,
both inanimate and animate, and this conception was also adopted by
Swedenborg. It was developed in the course of his foreign tour by studying
both the philosophy of Descartes and the writings of contemporary physicists
and anatomists.
Swedenborg s vieivs of the life-problem
His views of life are at first much the same as those we found in Hoffmann;
the body is a mechanism, to which is added a vegetative life-force, animus,
which consists of a fine material substance, and finally a higher soul, tnens.
^ After being ennobled he called himself Swedenborg, having previously borne the family
name of Swedberg.
l88 THE HISTORY OF BIOLOGY
But while Hoffmann leaves the latter to the metaphysicians, Swedenborg
becomes involved in speculations upon it; he holds that it likewise consists
of a fine material substance, which leaves the body at death and continues
to live in space; during life it receives mental impressions from the animus
and forms them into knowledge. But what interests him most deeply is the
question why knowledge is limited; like van Helmont he concludes that this
is due to the Fall, for before the Fall Adam was omniscient, and it now be-
came Swedenborg's aim to acquire this omniscience. He mainly sought to
gain it by studying the function of the brain and its relation to the life of
the soul.
Swedenborg's investigations of the brain really constitute the principal
part of his activities as a natural scientist. In this field he succeeded, by
brilliant comparison of conclusions drawn from the results of clinical post-
mortem examinations and from contemporary anatomical works — mainly
Malpighi's and Vieussens's researches referred to above • — in creating a
theory to explain the function of the central nervous system, which is far
superior to any that the anatomical specialists of his time were capable of
forming. Thus he localized the functions of the soul entirely in the cortex of
the great brain and was of the opinion that the corpuscula of the latter (the
pyramid-cells) discovered by Malpighi are connected by means of threads
with the various parts of the body and with one another, so that definite parts
of the body and definite parts of the cerebral cortex are conjoined to one
another and form the substructure for the functions of the soul ; it is through
this apparatus that the sensations are put into motion. This theory of the
brain, the value of which has been appreciated only in modern times, was,
however, made the basis for the most fantastic speculations on the soul,
which Swedenborg now believes to consist of a "fluidmn spirituostim,'' a
substance of exceptional fineness and directly derived from the eternal light.
It is impossible, owing to the existence of sin, for man during his earthly
life to come into contact with this supreme soul-substance, '' anima,'' which
possesses entirely ideal qualities, but he must be content w4th such lower
experiences as his mens and animus give him through the senses. Swedenborg
himself sought by way of desperate spiritual struggles to acquire that ideal
knowledge which man, in his view, had inaccessibly preserved within him,
but when he thought that he had attained his object after the vision men-
tioned above, his victory led merely to an initiation into the secrets of the
spiritual world, which has certainly conduced to the edification of the few
members of the Church he founded and of the far more numerous followers
of spiritualism, but which has proved absolutely useless to science and to
humanity at large, and which besides was the cause of the really splendid
contributions he made in the field of natural research being considerably
underestimated for a long time afterwards. It has been left to our own time
SEVENTEENTH AND EIGHTEENTH CENTURIES 189
to do him justice in this respect and to give him the place due to him in the
history of scientific research.
As will have been realized from the above, the theoretical speculations
to which reference has been made here led, on the whole, to poor results.
The general theories of life and its manifestations which were formed at the
period under discussion received a decidedly dogmatic stamp and became as
numerous as those who formulated them.
On those lines, therefore, it was impossible in the long run to achieve
any satisfactory results. Simultaneously with these efforts, however, there
appeared others which succeeded bettet in satisfying humanity's craving for
knowledge, and which during the immediately succeeding period won a
very large number of adherents — those works which comprised a systematic
description and classification of living creatures on earth. To these, then, we
shall now proceed.
CHAPTER VI
THE DEVELOPMENT OF SYSTEMATIC CLASSIFICATION
BEFORE LINN^US
Primitive systematic categories of animals and plants
AS LONG AS man's KNOWLEDGE OF NATURE is limited to what he can ob-
serve in his immediate vicinity, he has little difficulty in controlling
L the objects of his knowledge, but when his range of vision widens,
there arises the irresistible need for combining the individual objects that
have been observed under general expressions, which serve to fix the knowl-
edge of them and to impart it to others, "since no language would suffice to
denote everything individually, and since in a language which did so, no
understanding, no common knowledge, nor retention of such an infinity of
terms would be possible" (F. A. Lange). Those categories in which natural
objects are thus grouped by the most primitive peoples, out of sheer practical
necessity, are naturally based on such qualities in animals and plants as well
as the inanimate things that are observed as are easily comprehended, strik-
ing to the eye, and of special importance to the observers, and such terms are
also used and invented even today among civilized peoples by all those who
are concerned with nature in a purely practical way. On the other hand, a
grouping of natural objects based on scientific principles has taken a long
time to develop. In this respect the ancient Greek natural philosophy was
content with the primitive popular nomenclature. Practically the first to
devote scientific study to these groupings were, as far as we know, Plato
and Aristotle. From Plato originates grouping in species and genera — that
is to say, laterally arranged and superordinated terms — and his school still
further extended this grouping of terms: the dichotomical determination-
tables which even today play such an important part in plant and animal
systematization originate from his school. But the further this grouping of
terms went on, the more abstract became the result; the higher one came in
the series of terms arranged one above another, the further away has one
come from the things which one started from. This is a fact which the
biological systematicians have not always realized; the practical advantage
of systematic categories has led to the zoologist's and the botanist's for-
getting how artificial their system has really become.
System of Aristotle
In this direction Aristotle did not go beyond what Plato had initiated; in his
biological works there are, as is well known, only two systematical terms:
190
K
O
I— I
a,
<
o
u
jtf^p^^'--
w
Pi
o
o
:^
o
h-I
2
>
o
I— (
o
SEVENTEENTH AND EIGHTEENTH CENTURIES 191
eidos or species, and gems, the family, in which are included all combinations
of forms which come above the notion of species. Nor indeed has he given us
any really worked-out system; the animal system which is counted for his
has been compiled by others from his writings. His knowledge of forms was
also so slight that there seems to have been no difficulty in following the
simple grouping which he employed. As a matter of fact, during the centuries
that followed there was no need for a more detailed classification; the ani-
mals and plants which became known in late antiquity and the Middle Ages
were not so numerous that they could not be covered by the Aristotelean
natural philosophy. It was not until the great geographical discoveries of
the sixteenth and seventeenth centuries introduced the knowledge of a great
number of new life-forms that it was an inevitable necessity to widen the
biological classification if the material collected was not to accumulate
into an absolutely intractable mass.
The classification of plants especially demanded revision and expansion.
Actually it was long after zoology had done so that botany attained the
rank of an independent science. In antiquity and the Middle Ages botanical
knowledge was essentially supplementary to pharmacology. Aristotle's
botanical writings are, except for a few fragments, entirely lost. His disciple
Theophrastus' great work on plants was adopted by later writers as a model;
in it he thoroughly discusses the difference between plants and animals,
higher plants and higher animals being exclusively compared and the com-
parison developing into abstract and fruitless speculations. The old primitive
division into herbs, bushes, and trees is the only one to be found here. Be-
sides Theophrastus' work there was during classical antiquity a purely
pharmacological account of plants which was very celebrated and which
was ascribed to a philosopher named Dioscorides, whose character and period
are unknown (he probably lived at the beginning of the Christian era); it
was on Theophrastus and him that Pliny based the account of plants which
is included in his great Natural History. In the Middle Ages these writings,
which were believed to contain all the plants in existence, were closely stud-
ied and commented upon; attempts to find the plants from central Europe
in these works, which applied only to the Mediterranean countries, led to
the most absurd speculations. Only some few Arabian authors ventured
through all this long period to describe new plants. It was not until the
Renaissance that a change took place in this respect. One pioneer in this
field was Otto Brunfels, born, probably in 1488, in south Germany. In his
youth he was a monk; then he became a Lutheran and a schoolmaster at
Mainz; he died at Berne in 1534. He published an important work entitled
Herbarum viva eicones, which inspired Linnasus to call him the father of bot-
any. In this work, which was illustrated with excellent woodcuts, Brunfels
describes all the plants he knows. In his botanical descriptions he still partly
I9i THE HISTORY OF BIOLOGY
takes his stand by the old point of view; he begins each description with a list
of names in different languages, followed by an account of what ancient
authors have said of the plant in question; finally he gives his own "judg-
ment" on the plant and ends with a statement as to its "powers." Compared
with Gesner's exposition of the individual forms of animals (Part I, p. 93),
this is certainly clumsy, but as being the first of its kind the work at any
rate deserves respect. There is no system in it whatever; the book begins
with Plantago, plantain, "because it is common and because more than any
other plant it bears witness to God's omnipotence."
Thus it was at all events the medicinal powers of plants which most in-
terested Brunfels, and the same is true of his numerous successors in the six-
teenth century. The most interesting of these is Leonard Fuchs (1501-66),
who after working at humanistic studies under Catholic guidance went over
to Protestantism, devoted himself to medicine, and finally became professor
at Tubingen. His important botanical work Historia Stirpum, profusely and
beautifully illustrated, was published in 1541. Its chief interest lies in the
fact that he gives a list of all the terms he uses: an enumeration followed by
short descriptions of the names of the different parts of plants. Curiously
enough, the word "flower" is entirely absent. His description of individual
plants, as compared with Brunfels's, indicates an important advance; of
every plant an account is given of the (i) form, (2.) habitat, (3) season
(when it should be collected), (4) "temperament," (5) powers. It is only
under the last heading that the views of the ancient authorities are referred
to. Occasionally also the author, after the fashion of Aristotle, differen-
tiates to some extent between species and genus.
Cesalpino's -plant-system
The first to deal with botany as a truly independent science, however, was
Andrea Cesalpino (15 19-1603). His life was described in the first section
(Part I, p. 113), as also his general scientific point of view — strict Aristo-
teleanism. His great work on botany, De Plantis, is based on the same sys-
tem. Not only the fundamental ideas, but even the actual formal treatment
of the subject is entirely on the Aristotelean model: exhaustive comparative
analysis of the forms, concisely worded theoretical definitions, and, based
on these, abstract conclusions, without any idea of such practical utility as
was the main point with the old herbalists of the type of Brunfels. He begins
a definition of the difference between plants and animals in the true Aris-
totelean style: plants feed, grow, and produce offspring, but lack the sensi-
bility and motion of animals and therefore also need smaller organs than
animals. Then follows a comparison between vegetable and animal organs,
which, owing to its abstract one-sidedness, leads to curious results: the
alimental organs of plants are the roots; thus these correspond to the stomach
and intestinal canal in animals. Stalk and stem produce the fruit; thus they
SEVENTEENTH AND EIGHTEENTH CENTURIES 1 93
belong to the reproductive system. The plant is composed of several layers:
bark, liber, wood, pith; of these the pith is the innermost and thus corre-
sponds to the intestines of animals and is physiologically the most important.
He is at much pains to discover which part of the plant corresponds to the
heart in animals — we have previously pointed out (Part I, p. 1 13) the great
importance which Cesalpino attaches to the heart as the centre of the body
and of life. Finally, the plant's centre of life is found to be the collar of the
root — the place where the stem and the root system join. Thence extend
the vessels of the plant, of which the lacteal vessels especially are observed
and compared with the veins in animals. Propagation by means of cuttings
shows, however, that the central point of the plant is not as absolute as that
of the animal; with true Aristotelean terminology it is maintained that this
central point "actu" (actually) is in the root-collar, but "potentia" (po-
tentially) can be everywhere. Cesalpino, moreover, is particularly interested
in the fruits of plants, in which he sees the equivalent of the animal embryo;
the function of the leaves is to protect the fruits, and the flower-petals are
modified foils — an idea which was later adopted by Goethe. But Cesalpino
does not admit the existence of sex in plants: the fruit is formed from buds
and these again are produced out of the pith and the liber; the pith, which is
the most vital part of the plant, provides the actual ovule, and the liber gives
rise to the flower-leaf. Different kinds of fruits are carefully analysed and
the plants are classified in accordance therewith, though the traditional
division into trees, shrubs, half-shrubs, and herbs is retained as the main
division. These four categories are then divided in their turn, according to
the nature of the fruit, into a number of subdivisions. Cesalpino, however,
like Aristotle, makes no summary of his system, not even in the form of
chapter headings; nor is there any special systematic nomenclature. The
mulberry-tree, the hazel-bush, and other fruit-trees are thus described each
by itself; nevertheless, there sometimes occur divisions into lower categories
than those named: of the carrot. Caucus, for instance, three forms are men-
tioned, Creticus, Montanus, Campestris, a division which has the character
of a determination of species, or rather of variety. Nevertheless, these and
other categories occurring in Cesalpino are not sharply defined; he was un-
doubtedly more concerned with anatomical and physiological than with
systematic problems.
Cesalpino's system, in spite of its deficiencies, is the first to have been
really based on the comparative study of forms; in this connexion Linnasus,
who made a summary of it, expresses the opinion that Cesalpino is the first
to lay down a definite basis for plant classification. In later times, however,
this basis has been regarded as artificial, since it rests merely upon the con-
sideration of one single organ, and in contrast thereto have been adduced
contributions to a natural system of classification made by certain of the old.
194 THE HISTORY OF BIOLOGY
and otherwise not particularly systematic, herbalists. Independently of
Cesalpino, plant classification was actually developed in a new direction
through Caspar Bauhin. He was born at Basel in 1550 and studied medicine
and botany, as did also an elder brother, under the above-mentioned Fuchs
at Tubingen. He afterwards worked for a number of years as a professor in
Basel, until his death, in 1614. His chief botanical works, Prodromus and
Pinax theatri botanici, constitute the first attempts at a critical compilation
of all the then known scientific names and descriptions of plants.
Bauhin s system
Bauhin is entirely independent of Cesalpino; he bases his principles on his
master Fuchs and those like him, the semi-medical herbalists of the sixteenth
century. But he differs from the latter in his keen eye for the natural affinity
of plants; he groups together such plants as resemble one another generally
in their external form and discusses them in order, starting with those he
considers the most primitive: the Graminaceas, then the Liliaceas, the
Zingiberaceas, after which the dicotyledons, and finally shrubs and trees.
These groups are, however, neither characterized nor given names. Only the
individual plants are described, which are combined under one genus-name,
after which they are characterized in respect of all the forms that belong to
each one of those names. These diagnoses are brief and concise and are ac-
companied by short accounts of earlier authors' statements on each plant.
On the other hand, the actual genus-names are not in any way characterized,
any more than the larger groups mentioned above; there is therefore no justi-
fication for the assertion that is sometimes made that Bauhin clearly grasped
the contrast between genus and species. With greater reason he has been
called the originator of natural plant classification based on the common
likeness between the plant forms, as opposed to the artificial systematization
founded by Cesalpino, which is based on an individual organic system —
a contrast that has proved of great significance in botany, whereas in zoology
it has not been of such consequence. And above all as a critic of earlier botani-
cal literature Bauhin carried out a work of lasting value.
Joachim Jung, generally called Jungius, holds a peculiar position
amongst the botanists of the seventeenth century. Born at Liibeck in 1587,
he became, while still young, professor in mathematics at Giessen, but soon
relinquished his appointment, and thereafter, for more than ten years, he
lived a somewhat restless life, until in 162.8 he became rector of a gymnasium
in Hamburg. He displayed extraordinarily keen and many-sided activity
both as a scientist and as a tutor, but eventually he came to work in rather
difficult circumstances, partly owing to quarrels with the Hamburg priests,
who accused him of heresy.' For these and other reasons most of what he
' In the course of his education in Greek, Jung had studied, besides the New Testament,
profane classical authorsi when challenged on this point, he defended himself by saying that
SEVENTEENTH AND EIGHTEENTH CENTURIES 195
wrote remained unprinted and was partially dispersed after his death (in
1657). Some few treatises were published by his pupils, among them one
entitled Isagoge pbytoscopica (^Handbook of Botanical Study). This work, com-
prising a volume of forty-six quarto pages, must be regarded as one of the
pioneer works in botany. It gives a concentrated account of the theory of
botany, under the obvious influence of Cesalpino's, but without the latter's
profitless Aristotelean speculations; to begin with, the plant is characterized
as such, after which an account is given of the various organs, each of which
is briefly diagnosed in a manner that is striking, though abstract. "A leaf
is that which stretches out from its place of attachment in height and length
so that the surfaces of the third dimension are dissimilar to one another; it
is the leaf's inner surface that is differentiated from the outside." — The
whole exposition, with its concise, vigorous sentences and its analyses of
different parts of the plant drawn up in tabular form, is more reminiscent
of Linn^eus's work than that of any other of the early botanists. Linnsus,
in fact, mentions Jung as his precursor as far as the drawing up of rules for
the description of flowers is concerned and actually took up the characteristic
description of plant-organs at the point where Jung had finished and certainly
brought it up to a far higher standard.
One who in his time was of considerable importance as a classifier of
plants was Augustus Quirinus Rivinus (i65Z-i7i3). Born in Leipzig of a
family of scholars, which really bore the name of Bachmann, he studied
medicine in his native town, ultimately becoming a professor there. He was
a many-sided scholar, working in widely differing spheres; his chief fame,
however, rests on his great botanical work Ordo flantarujn, which he pub-
lished in two large folio volumes, illustrated with fine copper engravings,
entirely at his own expense. He was the first to insist that the old division
into trees, bushes, and herbs should be done away with; in its place he would
classify plants exclusively according to their corolla, and he thus created an
artificial system, which, however, was not very practical. He likewise urged
the adoption of a simplified nomenclature for the plants themselves, but
in this, too, his criticism of the old system was more successful than his
attempts at reform.
A far greater service to classification was rendered by Joseph Pitton de
TouRNEFORT, at just about the same time. He was born at Aix in the south
of France in 1656 and was destined by his father for the priesthood — much
against his will. When his father died, therefore, he gave up theology and
the latter wrote purer Greek than that of the New Testament, whereupon the priests in Hamburg
and theologians at Wittenberg accused him of blasphemy, because he had reproached the Holy
Spirit, which had inspired the words of the Bible, with a deficient knowledge of languages.
Jung had to abridge his school education, but, thanks probably to his high reputation, escaped
the sentence of excommunication with which he was threatened.
1 96 THE HISTORY OF BIOLOGY
applied himself to botany, which had always interested him. In order to be
able to earn his living he started by taking the degree of doctor of medicine.
His botanical works soon gained him a wide reputation; he was appointed
professor at the Jardin des Plantes in Paris and had an opportunity of making
many long journeys for research purposes. He died in 1708 as the result of an
accident.
In the introduction to the important botanical work in which he sum-
marized the results of his research activities he expounds his principles of
plant classification. He defines the plant as an organic body, which always
possesses roots, practically always seeds, and nearly always stalk, leaves,
and flowers. He bases his ideas of the structure of plants on Cesalpino and
Malpighi. When later it comes to classifying and giving characters to plants,
he maintains, under the manifest influence of Cesalpino, that only the flowers
and fruits can come into question; he seeks far and wide for proofs as to why
root, stalk, and leaf do not provide reliable characters. In particular, the
plant genera should be based on similarities in the structure of the flowers
and fruits, but as the same genus includes forms whose remaining parts are
different, so the genera must in their turn be divided into sub-categories.
Tournefort pays great attention to his description of the genera, and his
diagnoses of them are often so striking that subsequent systematicians, up
to our own time, have been able to accept them, though they are only based
on the characteristics of flower and fruit; on the other hand, the "species"
into which the genera are divided are mentioned with only a few words
regarding the form of the stalk and the leaf, without any further description.
His method of procedure is thus the exact opposite of Bauhin's. But over and
above this, Tournefort works out for the first time a systematic classification
of categories higher than the genera — that is to say, he divides the plants
into a number of classes, which again are severally divided into sections;
each of these is characterized in a few words, but is not given a name. The
characters of these higher categories are derived from the peculiarities of the
flower; several categories of flowers which still to some extent hold good
today are determined by him: with and without corolla, with or without a
gamopetalous corolla, and, again, cruciform, Ungulate, and other flower-
forms. The division into herbs, bushes, and trees abolished by Rivinus he
himself, however, was never able entirely to reject; his system comprises
seventeen classes of herbs and five classes of bushes and trees. With regard
to anatomy and physiology Tournefort has not much to ofi'er that is new; in
the course of his journeys he had observed the artificial fertilization of date-
palms practised in very remote periods and already described by Theophras-
tus. They are, as is well known, both male and female, and the cultivators
facilitate fertilization by suspending male clusters over the females, but
Tournefort is unable to derive any theoretical conclusions of importance
SEVENTEENTH aND EIGHTEENTH CENTURIES 1 97
from the fact. It was left to another scientist, Camerarius, to prove the sexu-
ality of plants.
Sexuality of plants
It was known of old that in certain plants the individuals are of two differ-
ent kinds, both of which must concur before any reproduction by means of
fertilization can take place. The classical example of this, known to all the
natural philosophers of antiquity, is, as mentioned above, the date-palm,
the fruitful specimens of which have been quite correctly called, by the
peoples who cultivate them, females, while those that are required for
fertilization have been called males. But other plants of the same kind have
also been known since ancient times, though many plants that resembled one
another, but were differentiated by varying size and development were taken
for females and males. A well-known instance of this was that of the two
ferns Filix mas and Filix femina, which are still retained as names of species
in two different fern-genera. But these ideas mostly belong to popular belief;
scientists, both those of the classical period and, on their authority, those
of the sixteenth and seventeenth centuries, denied, or at any rate overlooked,
the existence of sexuality in plants, owing mostly to the fact that the great
majority of plants are hermaphrodites. When no difference can be found in
the male and female specimen, what is the use of assuming sexual reproduc-
tion? Grew was the first to believe that plants reproduce themselves sex-
ually, "like snails" (these are, of course, also hermaphrodites). His opinion
in this case, however, was based mostly upon theoretical speculation, and,
as a rule, such speculations are, of course, less convincing than direct obser-
vation. The scientist who proved the sexuality of plants as the result of
convincing experiments was Rudolph Jacob Camerarius (1665-17^1). He
belonged to an old scholarly family, known since the Renaissance period,
which had originally been called Cammerer, and he worked throughout his
life at Tubingen, where he was for many years professor of medicine. He
generally recorded the results of his work in small articles, frequently writ-
ten, according to the custom of the period, in the form of letters to other
scholars. The essay which alone justifies the mention of his name in a history
of biology is a "Letter on the Sex of Plants," dated 1694. In this article he
gives an exhaustive account of all the ancient authorities' ideas of the re-
production of plants and of the parts of flowers; he himself arrives at the
conclusion that the pollen is the male, and the ovary is the female, element
and discusses in connexion therewith a number of theories on sexuality and
fertilization in general, without, however, contributing anything of special
value from a theoretical point of view. Of all the greater significance are
the experiments by which he proves his theory of the sexual properties of
plants. He cultivated for this purpose a fairly large number of both monoe-
cious and dioecious plants and found that i^ the male flowers are picked off
198 THE HISTORY OF BIOLOGY
in time, there will be no fruit, while fruit will certainly develop if the pistils
of the female flowers are provided with pollen. These proofs had undoubtedly
a convincing effect, if not on all his contemporaries, at any rate on succeeding
ages. Linnasus in particular has acknowledged the contribution he made to
the development of plant physiology.
Animal system neglected
While, then, during the first two centuries of the new era plant classification
was splendidly reorganized, during the same period animal classification on
the whole made no progress. The zoography of the Renaissance period has
already been described (Part I, pp. 31-8); it was, generally speaking, not very
systematic; in the best event one adhered to Aristotle, and in the latter's
Historia animalium zoology had, in fact, an old and sound foundation, which
contemporary botany lacked — a careful comparison, based on unique pow-
ers of observation and sense of form, between the individual animal forms,
the value of which is manifest from the fact that most of the groups into
which animals are there divided still hold good in the present system of
classification. Particularly in regard to vertebrate animals, which have for
obvious reasons been of primary interest to humanity, Aristotle had, as has
already been pointed out, a keen eye for the natural affinity between the
different forms, which is based upon agreement in the general structure and
functions of the body. Thus there was opened up to animal biology during
this period an important and fruitful field for research in the anatomical
and physiological sphere-, and this, again, caused the comparison between
the life-forms in the animal kingdom to receive a different character from
that between the life-forms in the vegetable kingdom; in the former a com-
parison between internal organs, the complex structure of which it was pos-
sible to make out only after exhaustive investigations; in the latter, a study
for the most part of problems of the purely external form. In zoology, too,
however, it was absolutely necessary to develop form classification, mainly
owing to the fact that the different categories into which the known animal
world is divided required a more definite determination than that given it
by Aristotle and his successors. And this undoubtedly demanded co-opera-
tion between zoology and botany in order to find a common ground of com-
parison and valuation for all the forms in which life on earth manifests
itself. The very first to make an attempt to deal with vegetable and animal
classification on similar principles was Ray; the scientist who finally worked
out a uniform system for all living creatures was Linn^us.
John Ray was born in 16x7 or i6x8 at Black Notley, a village near Brain-
tree, in Essex. His father was a well-to-do blacksmith who could afford to
send his eldest son to college. In 1644 young Ray went up to Cambridge
and at first studied the classical languages and theology; but he was also
interested in mathematics and natural science. He gave lectures to the under-
SEVENTEENTH AND EIGHTEENTH CENTURIES 199
graduates on Greek and mathematics alternately, and was eventually or-
dained, after which he held many college offices. His university period was
not to last long, however; the reactionary Government of Charles II required
the English clergy to subscribe to an Act of Uniformity drawn up with a
view to suppressing liberty of conscience; and Ray was one of those who
preferred to give up office rather than to submit. It thus came about that,
like so many of England's best scientists, he had to spend the greater part of
his life following the profession of a private scholar. This Ray was enabled to
do thanks to his connexion with Francis Willughby, a very wealthy young
man of noble family, who, eight years younger than Ray, had been a pupil
of his at Cambridge and was his constant companion throughout his life,
their friendship being based on a common interest in natural science. After
Ray's resignation the two friends went for a several years' tour through
Europe, during the course of which Ray applied himself especially to botany,
Willughby to zoology. Having returned home laden with collections, they
settled down in Willughby's country-house in order to work up the material
they had collected. In 1672., however, Willughby's death abruptly terminated
their collaboration; by his will he appointed Ray one of his executors and
left him sixty pounds a year for life, with the charge of educating his two
sons, for which purpose Ray remained for some years in his friend's family.
Having married, he finally settled down in his parents' cottage, which he
had inherited, and there for several decades he continued his researches,
universally respected in scientific circles in England and contented with his
lot in spite of his modest circumstances. He died in 1705, three daughters
surviving him.
Kay's Methodus plantarum
Ray's literary work was extensive and many-sided — sermons and religious
essays, handbooks on the classics, treatises on folk-lore, and, finally, the
works on natural science on which his fame entirely rests. The greatest of
these, in both volume and importance, is his Historia plantarum generalis,
a work of i, 860 closely printed folio pages, in which he summarized the entire
botanical knowledge of his time. At an earlier date he published a resume of
the system in which he arranged the plants in his Historia, under the title
of Methodus plantarum. This great history, which contains a systematic de-
scription of all the then known plants, starts with a general survey of the
nature and conditions of plants. He quotes Aristotle's principle as to the
division of the organs into simple and complex, similar and dissimilar. As
regards the various parts of the plants he bases his system on Jung's defini-
tions and terminology, which are regularly quoted, but are in each individual
case considerably extended and thoroughly investigated. He cites the plant
classification which Cesalpino originated, according to fruits and seeds, but
he points out that the form of leaves and other parts must also be taken into
XOO THE HISTORY OF BIOLOGY
account, so that plants which resemble one another are grouped together
although the seeds are different. Above all, he reminds us that nature never
makes any jumps; on the contrary, the extremes are connected by middle
forms, just as the zoophytes come between the vegetable and animal king-
doms. In regard to the anatomy of plants, in all essentials Ray follows Mal-
pighi; Grew's ideas on the sexuality of plants are also accepted, without,
however, being further developed; Ray was ignorant of Camerarius's obser-
vations. On the other hand, he describes the germination of plants, making
original observations of considerable value; the difference between plants
with one and those with two cotyledons was established by him. Ray
discussed the notion of species more thoroughly than any previous biolo-
gist. In his view, plants belong to the same species if they give rise through
their seed to a new plant similar to themselves, in the same way as bulls
and cows are the same species because in mating they produce creatures
which resemble themselves. The number of species is invariable, for God
rested on the seventh day from all his work — that is, from creating
new species. On the other hand, the different-coloured flowers in plants
should not be regarded as separate species, any more than the different-
coloured calves born of cows; in the former this is proved by the fact that
the colour variations are not reproduced through seed, but only through
cuttings. The invariability of species is, however, not absolute; plant species
can be varied through the "degeneration" of the seeds — thus it has cer-
tainly occurred that the seed of the cauliflower has produced leaf-cabbage
and that from the seed of frimula veris t?iajor has arisen primula fratensis in-
odora. Ray even includes in the discussion a number of ancient stories as to
grain's having degenerated into weed: wheat to Lolium and maize to other
kinds of weed. True, he doubts the truth of a number of these statements,
but he nevertheless believes the thing to be possible. This belief of his in
the variability of species has been cited as proof of an unprejudiced view, in
contrast to the theory that arose later as to the absolute constancy of species.
The examples quoted rather go to show clearly enough that Ray was unable
to rid himself of a certain amount of primitive superstition.
As far as actual classification was concerned, Ray retained the division
into herbs and trees, or, more correctly, herbaceous plants and ligneous
plants, maintaining that the latter are differentiated from the former by the
existence of winter buds — in actual fact, an incorrect assumption. In a later
edition of his Methodus, however, influenced by Rivinus, he abandoned this
division. Herbs are then divided into: tm-perfecta (fungi, alga;, lichens, and
corals) and perfect a (plants bearing flowers, which are again divided intu
those having tw^o and those having one cotyledon). The sub-groups under
these categories are numerous, some natural and well characterized, others
composed of all sorts of plants, massed together owing to some purely ac-
SEVENTEENTH AND EIGHTEENTH CENTURIES 2.0I
cidental character. Trees are also divided according to the number of cotyle-
dons, and then again into sub-groups. The actual genera are mentioned by
one name and are described with a short diagnosis; the species into which
they are divided are characterized in a few words, followed by a more de-
tailed description. Ray was thus the first to describe both genus and species
at the same time.
Ray's zoological system
As a zoologist Ray left no comprehensive work corresponding to the bo-
tanical work above mentioned. During his period of coUabc ration with
Willughby the latter took over the zoological side, and after his premature
death Ray published in his name a couple of works on birds and fishes; how
much in th .se works originated from the one or from the other of the two
friends it is not easy to decide. In his own name, on the other hand, Ray
promulgated two zoological works: a survey of the quadrupeds and reptiles,
and a work on insects. The former of these, a small cctavo vclume, is his
most important contribution to the knowledge of the animal kingdom. He
begins with some general reflections on the characteristics of animals; he
defines the animal as a body having life and powers of perception and of
independent motion, and he then discusses Descartes's assertion that animals
lack sensibility, the incorrectness of which is proved. As regards the repro-
duction of animals, he denies spontaneous generation and then deals with
the theories of epigenesis and preformation, ovism and animalculism, with-
out making any very important contributions in that connexion. The theory
of fabulous creatures, which had always up to then been included on the
authority of classical authors, is examined and entirely exploded. In regard
to systematic classification, which comprises the greater part of the work,
Ray follows Aristotle in essentials, and this for good reasons, since the lat-
ter's division of the quadrupeds is on the w^hole both natural and well
founded. Ray, however, did not venture to follow up the consequences of
the comparative anatomical method which Aristotle founded; like the latter
he refers whales to the fishes, although he is quite well aware of their closer
anatomical affinity with the mammals. On the other hand, in certain respects
Ray goes deeper into the characteristics of the individual animal groups
which he adopted; above all, he takes account of the structure of the circula-
tory organs and on this basis divides animals first of all into sanguiferous
and bloodless — he quite realized that the last-mentioned group possesses
blood of a kind, though colourless, but he prefers to retain the Aristotelean
nomenclature. The sanguiferous animals are divided into those that breathe
wit'i lungs and those that breathe with gills; those provided with lungs are
again divided into animals having two heart- ventricles and animals with only
one To the last belong oviparous quadrupeds and reptiles; the first is divided
into oviparous (birds) and viviparous (partly land animals — mammals —
2.0-L THE HISTORY OF BIOLOGY
partly aquatic animals — whales). Land animals are also characterized by
their hairy covering, as the result of which the manatee, which lives in the
water, can be included among them. The bloodless animals are divided into
small (the insects) and large (molluscs, crayfish, crustaceans). In examining
this system we may pass over the ' ' bloodless ' ' animals, of which Ray himself
only studied the insects; aquatic animals were, on the whole, of no interest
to him, and as, moreover, Willughby had devoted himself to birds and fishes,
there remained only the quadrupeds, which, as mentioned above, formed
the subject of his principal zoological work. The hairy quadrupeds are di-
vided into Ungulata or hoofed animals, and Unguiculata or clawed animals.
Among the former are reckoned one-hoofed (equine), pair-hoofed (Rumi-
nantia and swine), and multi-hoofed (rhinoceros and hippopotamus).
Amongst the Unguiculata are included pair-clawed (the camel) and multi-
clawed; (i) with claws grown together (the elephant); (x) with separate
claws, of which there are: flat claws (apes) and narrow claws (carnivorous
animals and Rodentia). Moreover, a number of mammals are classified as
"anomalous" — namely, the hedgehog, the molcj the shrew-mouse, the
armadillo, the sloth, and the bat. The oviparous quadrupeds are finally
divided into frogs (including tortoises), lizards, and snakes. In this system
each genus is then characterized with a diagnosis — for instance, the genus
Ovis, the genus Martes — and the species of the genera are likewise given
each a separate diagnosis. On the other hand, the genera of frogs, lizards,
and snakes are not diagnosed, only a common characteristic being named,
followed by diagnosis of the species.
To measure Ray's work as a systematician by modern standards would
naturally be entirely unhistorical, but his system can by no means bear com-
parison even with that of Linnasus. And yet for his age it constitutes an
extraordinary advance, primarily in that he clearly realized the difference
between species and genus, secondly on account of his possessing what was
undeniably — in comparison with his predecessors — an extremely keen eye
for the similarities on which the assumption of affinity in its wider sense
may be based; several of his larger groups, both in the vegetable and in the
animal kingdom, are "natural" in the best sense of the word. In the sphere
of botany, also, the difference discovered by him between mono- and dicoty-
ledons is of essential importance. On the other hand, several of the sub-
divisions which he formed are highly artificial, as will be clearly seen from
a glance at his division of mammals, according to claws and nails. And in
any case he established no common systematic categories to cover all living
creatures. The one who by doing so paved the way for a completely uniform
conception of life-form on this earth was Linnasus, the founder of modern
plant and animal classification.
CHAPTER VII
LINN^US AND HIS PUPILS
Linnaus's life and ivork
NILS Ingemarsson was a peasant lad from Sunnerbo, in the province
of Smaland in Sweden, who was destined for the priesthood. When at
school, not having previously had any family name, as was the case
with the country people in general in Sweden, he adopted the name of Lin-
nasus, after a mighty linden-tree growing near his home, which was regarded
by the country folk as a sort of sacred tree. After a long period of study at
Lund University — frequently interrupted, owing to his poverty — he was
ordained priest in 1704 at the age of thirty, and two years later he was ap-
pointed curate at Rashult. At the same time he married Christina Brodersonia,
daughter of the Vicar of Stenbrohult. Some years later he succeeded his father-
in-law as vicar of that place. While following his vocation he also devoted
himself with keen enthusiasm to horticulture and the study of herbs; in his
large garden grew many a herb that was not to be found in his neighbours'
gardens and with the peculiar properties of which he was well acquainted.
The eldest of his large family was a son, Carl, born on the 2.3rd May 1707.
Even in his earliest childhood Carl displayed the same keen interest in botany
that his father had done; his greatest joy was to work in the small garden
he had had laid out and there to cultivate as many remarkable plants as
possible. At his school, at Vaxio, however, he was, as he himself relates,
far from happy; "crude schoolmasters in a crude manner gave the children
a mind for sciences enough to make their hair stand on end." In humanistics,
which at that time were the most important, he likewise made but little
progress, but he was all the more successful in the physical-mathematical
subjects. His teacher in physics, Rothman, quickly recognizing his great gift
for natural science, gave him Boerhaave's and Tournefort's works to read
and urged Carl's family to accept his plan to devote himself to medicine
instead of studying for the priesthood. In 172.7 he became an undergraduate
at Lund, where he found a paternal friend in Stobasus, professor of medicine.
On the advice of Rothman, however, he removed for the next academical
year to Upsala, where the medical teaching was considered to be of a higher
standard according to the requirements of the age, which, however, is not
saying very much. Linnasus had for the most part to carry on his studies by
himself. During his first term at Upsala he lived in dire want, but he soon
103
2.04 THE HISTORY OF BIOLOGY
succeeded in procuring patrons there: the dean, Celsius, who was likewise
interested in botany, took him into his family and undertook to procure
him further advancement. Even as a young student Linnaeus had always
shown that capacity which never left him throughout his life, of exciting
the admiration and sympathy of those he met who possessed interests simi-
lar to his own — a quality based on the keenness with which he himself
embraced the work he had made his own. When once he had acquired friends
at the University he gained one success after another. Though not yet a
graduate, he obtained permission to lecture on botany and he used to attract
large audiences. He received a number of grants, and with the aid of public
funds he made journeys of exploration to the Lapp district and Dalecarlia,
in the course of which he collected material for research consisting not only
of natural objects, but also of human customs and habits. During the latter
expedition he made the acquaintance of his future wife, daughter of the
wealthy town-physician of Falun, Morasus. In order to secure further ad-
vancement in the career he had chosen, Linnasus had to obtain the degree of
doctor of medicine, but there was no such degree in Sweden at that time.
He accordingly made a journey, with the financial support of his future
father-in-law, to Holland, where at the small university of Harderwijk,
which never attained to a very high standard of scholarship, he took his
doctor's degree in a couple of weeks. By that time, however, the money he
had brought with him had become exhausted and Linnasus had no other
resource than to chance his luck elsewhere. He accordingly went, in company
with a fellow-countryman, to Amsterdam and thence to Leyden. There he
became acquainted with several scientists and people interested in science,
chief of whom was Boerhaave, who treated him with paternal kindness.
With the assistance of one or two patrons Linnasus was able to print his
most epoch-making work, the Sy sterna ncitura, which he had already begun
in Sweden and which brought him immediate fame. He then spent three
years visiting the principal centres of learning in Holland, publishing one
work after another with marvellous rapidity, supported by patrons and often
almost persuaded to settle in Holland for good. He longed to return home,
however, and after paying visits to both England and France, he returned to
Sweden with a European reputation, but without any very brilliant prospects
for the future. He succeeded, though with some difficulty at first, in making
a living as a physician in Stockholm, until in 1741 he won the position for
which he had striven so long — the professorship of botany at Upsala.
During his Stockholm period he had taken part in the founding of the Acad-
emy of Science and had been its first principal; at Upsala, from the day of
his arrival, he became the foremost member of the University. His time and
capacity for work sufficed for everything — for his teaching, which went
on summer and winter, before ever-increasing audiences, both of Swedes and
SEVENTEENTH AND EIGHTEENTH CENTURIES XO<^
foreigners; for the reorganization of the botanical garden (which existed
in Rudbeck's time, but which had now fallen into decay), making it one
of the finest in Europe; for the production of extraordinarily fine scientific
works and an extensive correspondence. As a founder of schools and an organ-
izer of work he has had few equals in the history of biology. Every year
he sent out pupils on research expeditions, whose collections and observa-
tions were afterwards worked up under the master's own guidance. He him-
self was acknowledged throughout the whole civilized world as an authority
on natural-scientific questions, his advice being sought by governments as
well as private individuals. His native country also learnt to appreciate him;
he received several high honours; among other things he was ennobled and
took the name of von Linne.
The climax of Linnxus's greatness falls within the period of the seven-
teen-fifties; then he published the last of his great works, and then, too,
he received his highest honours. The quarter of a century of life that still
remained to him was a period of decline. The hardships suffered in his youth
and the cares of his m^turer years had undermined his health. By the begin-
ning of the fifties he had already become seriously ill, but he still managed
to work during the succeeding decades — in part producing results of con-
siderable importance — although his powers of movement ■ began to fail.
During the seventies, however, he was subject to repeated paralytic strokes,
which dulled his intelligence and finally paralysed him entirely. In 1778
death brought release.
In 1763 Linnaeus had taken a step which was certainly the most unfor-
tunate he ever took in his life; he had obtained from the Government the
right to recommend his successor and he appointed his only son, Carl von
Linne the younger, who thus at the age of twenty-two became aspirant to
the professorship, possessed no brilliant gifts and had never passed any tests
of scholarship. Although his promotion was by no means so ridiculous in
the eyes of his contemporaries as it would have been in modern times — it
was quite usual for people to purchase a "survivance" to an official post
similar to that which young Linne obtained on account of his father's serv-
ices — nevertheless this step had the most unfortunate consequences. The
feelings entertained by the large crowd of far more competent pupils were
naturally very bitter and were enhanced the more the worthlessness of young
Linne's character manifested itself, as it unfortunately did very soon, no
doubt hastened on by his unmerited promotion. And, to make matters worse,
it caused also a division in the Linne family. On his father's death the son
laid claim to his collections, which his mother and sisters, supported by a
will, refused to allow. The quarrel was finally settled by arrangement, and
shortly afterwards Linne the younger died, at the early age of forty-two,
after a life which brought little honour to the name he bore and which died
io6 THE HISTORY OF BIOLOGY
out with him. The unlucky position into which he got himself as regards
both his family and his colleagues was also no doubt responsible for the
sale (ignominious indeed for his country) of the collections of Linnasus —
his herbarium, library, and correspondence — to England, where they are
still preserved by the Linnean Society, which was founded for that purpose.
His fame
LiNN^us has in the course of years been very differently judged. Already
in his youth he had been hailed by his contemporaries as the " princeps botan-
icorum,'' a title that he succeeded in holding, not only throughout his life,
but long after his death. But the reverse came in connexion with the accept-
ance of the descent theory in the middle of last century, for the opponents
of this doctrine quoted Linnaeus as their chief authority, and that not only
on scientific grounds but also from motives which lay far removed from all
that natural science means: his primitive Christian piety was thrust into
the breach by religious and social conservatism against the "unbelief" of
the new biology. It was naturally inconceivable that in such circumstances
Linnasus and his works should be judged with impartiality; in the eyes of
many he became simply the arch-enemy of the new science, and the judg-
ments passed on him at the time were often not only spiteful, but also utterly
absurd. Towards the close of the century, however, a calmer atmosphere
prevailed, as was clearly manifested when in 1907 the bicentenary of Lin-
naeus's birth was celebrated by the entire civilized world as a red-letter
day in the annals of human culture.
Linnasus is universally reckoned among the examples of early scientific
maturity, and it is true that by the time he had reached about his twenty-
fifth year, he had already fully worked out the principles on which his sub-
sequent work rests. Less remarked has been the steady development which
he underwent so long as he was generally capable of working; the Linnasus
whom we meet in the first edition of the Systetna natura and writings contem-
porary therewith is not in the least the same person as the one who composed
the final editions of that work. This may to some extent explain why such
contradictory judgments have been passed on him; the one has sought sup-
port for its opinion of him in the work of his youth, the other in that of
his old age.
His general conceptions of nature
If on the basis of Linnasus's writings we were to try to form an opinion as
to his general conception of nature, we should soon discover that he never
formulated any elaborate theory of the phenomena of life in their entirety,
such as Hoffmann, Stahl, and Boerhaave did, each in his own way. In the
works of his youth there appears only a naively popular conception of nature,
which, as a matter of fact, he retained, practically speaking, throughout
his life : nature is created by God to His honour and for the blessing of man-
SEVENTEENTH AND EIGHTEENTH CENTURIES 107
kind, and everything that happens happens at His command and under His
guidance. No other explanation of natural phenomena is looked for. How-
little Linnieus actually interested himself in the general scientific questions
that occupied the minds of his contemporaries is at once shown by the fact
that even in the twelfth edition of his Sy sterna natura he still lets the universe
consist of the ancient four elements, fire, air, water, and earth; he seems not
to have been aware of the fact that many decades earlier Stahl had published
his new theory of the process of combustion. On the other hand, in this and
in other works of his later years there occur a number of ideas contributing
to a mechanical explanation of life that are in striking contrast to his ro-
mantic piety. In the above-mentioned edition of Systema natura he defines
(on p. 1 5) animal life as a hydraulic machine which is kept going by an
ethereal-electric fire maintained by breathing ;i on the other hand there comes
in here the universally known, sublimely poetical description of God's om-
nipotence: how he saw the Eternal wherever he went, and how his brain
reeled when he saw traces of Him in everything, from the life of the minutest
creatures here on earth to the movements of the heavenly bodies, "which
are upheld in their empty nothingness by the first movement, the essence
of all things, the mainspring and director of all causes, the Lord and Master
of this world; should we call Him Fate, we should not be wrong, for every-
thing hangs upon His finger; should we call Him Nature, we should not be
wrong either, for all things have emanated from Him; should we call Him
Providence, we should likewise be right, for everything happens according
to His nod and His will." The strange, half-pantheistic conception of God
that is here apparent occurs in Seneca, whose Quastiones naturales Linnaeus
cites in this connexion and often elsewhere; besides which he quotes in the
work in question the Bible, Aristotle, Cesalpino, and van Helmont in support
of his theory of the universe and of life-phenomena. To Galileo's physics,
Newton's astronomy, and Stahl's chemistry, on the other hand, he has paid
no attention; at any rate, there are no quotations that would indicate his
having done so.
His gifts as a systematician
At the time when polemics were levelled at him, Linnasus was accused of
Aristoteleanism in a derogatory sense. This accusation may have a certain
amount of justification, but it is likely in all ages to be laid at the door of
everyone desirous of arranging things according to formal principles, and
that was what Linnasus desired, just as it was exactly what biology in his
time needed. Far from blaming him for it, therefore, posterity should, on
the contrary, be grateful to him for having, instead of working out specu-
^ Is it possible that the "fire-machine" constructed by Triewald, which Linnaeus saw in
his youth in the mine at Dannemora, may have been recalled to his mind and have given rise
to this curious definition?
Xo8 THE HISTORY OF BIOLOGY
latively some doubtless unproductive system of thought, devoted himself
entirely to ascertaining the relation of forms to one another — an investi-
gation which was so suited to his peculiar gifts. That in doing so he accepted
the old biblical conception of nature was as natural in his day as for a system-
atist of our own day to embrace the theory of descent without going closely
into the question of its justification. He was thereby able, unhindered by any
theoretical barriers, freely to develop and take advantage of that extraor-
dinary capacity for observing natural objects and summarizing his obser-
vations which was peculiar to him, and thus to establish the mastery over
research-material on which modern biology is based.
Linnxus was, as has already been mentioned, essentially autodidactic,
in so far as the education he received from others was highly deficient and
fragmentary. Nevertheless, the conditions under which he was trained for
his life's work were particularly suited to his natural genius. The powers
of observation which formed one of his most conspicuous characteristics had
received from his very earliest years in his father's garden, under his guidance
and under the influence of the love of the vegetable world imparted by him,
such stimulating exercise as to afford every opportunity for the full develop-
ment of his extraordinary sense of form. During his youth his teachers gave
him a knowledge of such biological literature as then existed, without at
the same time burdening him with any theories out of which he would
afterwards have had to work himself up, while at an early age he gained
that liberty of action which is the indispensable condition for anyone who,
in whatever sphere his work may lie, wishes to create something new. The
works of his youth, the small lists of plants which he drew up and which
were not printed until our own day, already give clear evidence of where
his chief interest lay; he enumerates the plants he collected in various places,
with observations as to their occurrence, and by means of them he tests the
various systems which he found amongst his predecessors, principally Tour-
nefort, but also Ray and Rivinus, without, however, finding any real satis-
faction in them. On the contrary, we find from his notes that he felt himself
called upon to reform the science of botany, which he considered to have
seriously degenerated. Thus he became aware, through a criticism in a jour-
nal, of Camerarius's discovery of sex in plants, which a French naturalist
had accepted, and he was so excited by the news that he at once devoted
himself to making a close study of the problem. He immediately realized
that in the hitherto neglected stamens and pistils one had to do with the
flower's most vital organs, and from that point of view alone their employ-
ment as a basis for systematic classification was justified. Thus arose his
sexual system, the first step towards the realization of the ambition he had
set himself to attain: a general system for natural objects. And at the same
time he made himself quite clear as to the principles on which such a system-
SEVENTEENTH AND EIGHTEENTH CENTURIES 109
atic classification would have to be worked out, the result being the second
of the important works written in his youth, Methodus plantariun, wherein
he presented most of the principles which have since then been the common
property of plant and animal classification. He first of all laid down an ex-
planation and a definition of the various parts of the plant, after the model of
Jung and Ray, whom, however, he far surpasses in the matter of precision,
both of observation and of expression. Further, he worked out in an incom-
parable manner the principles of nomenclature, synonymy, and character-
istics of the various categories of the system, all of which have since then
been the common property of all systematicians of any ability, but which
in his time reacted with all the overwhelming force of a novel idea. With
all these ideas partly written down, partly in his head, Linnaeus came to
Holland, and was able, under the unusually favourable conditions which
he enjoyed there, to make them fully available for research. But before
giving an account of those, the greatest works of his life, we must devote
a few words to a man with whom he closely collaborated and who made a
strong and lasting impression on him.
Peter Artedi was born in 1705 at Anundsjo, in northern Sweden, the
son of a priest named Arctxdius. He entered Upsala University in 17x4 and,
like Linnasus, had difficulty in obtaining his father's consent to his studying
medicine in preference to theology. It was natural science, however, that
chiefly attracted him, and in that field he too, like Linn^us, had for the
most part to study on his own. At the time when Linnasus came to Upsala,
Artedi was considered the most promising naturalist in the University and
there soon arose a firm friendship between them, resulting in a co-operation
which proved of great benefit to both. Artedi was especially interested in
zoology, chiefly in icthyology, while Linnaeus applied himself to botany,
so that there was no necessity for them to encroach upon one another's fields
of activity, but at the same time they could exchange ideas and observations.
In their characters, too, they were fortunate in being able to complement
one another; Linnasus was lively and enthusiastic, Artedi calm and critical.
Financially they were both in an equally bad way and they had recourse
to one another's assistance. In 1734 Artedi received a grant to enable him
to travel abroad and he went to London, where he studied zoology, mainly
icthyology. A year later he came to Amsterdam without resources and with-
out the slightest prospect of getting home. Linnasus, who had already ac-
quired some connexions in the city, introduced his friend to a wealthy
apothecary who possessed a large collection of fishes. This museum Artedi
was now commissioned to catalogue and was able at the same time to com-
plete a large work on fishes on which he had long been engaged. His career,
however, was short; one evening, upon returning from a visit to his bene-
factor, he fell into a canal and was drowned (autumn, 1735).
iio THE HISTORY OF BIOLOGY
The icthyology of Artedi
His work was published by Linnasus with the assistance of a Dutch patron.
It was probably in all essentials the work of Artedi, though Linnasus made
some additions here and there. The work purports to be a complete mono-
graph on fishes; the anatomical section, however, is of minor importance.
The chief interest lies in the presentation of the theory of the system, which
is incorporated in the part entitled " Philosophia ktbyologka.'' In this are
discussed with sharp criticism the various systematical categories. He starts,
after the model of Tournefort, with the genus, which is defined as a collec-
tion of species that, as regards the shape, position, number, and mutual
relation of the parts, agree with one another and differ from other genera.
The species, moreover, he bases, not like Ray and Linnasus on common ori-
gin, but on dissimilarity in the same genus in respect of some individual
part of the body, a principle the weakness of which in comparison with
Linnicus's becomes at once apparent. As higher categories he adduces classes
and orders; the classes should be "natural" — that is, be based upon agree-
ment in several essential parts and not upon unessential factors, such as oc-
currence, size, and the like. Fishes form one such "natural" class, owing
to the shape of their body and their fins, whales nevertheless still being
counted in the "natural" class. The orders into which the class is divided
are on the whole the same as those still in use today — a proof of Artedi 's
systematical acumen; selachians, acanthopterygian and malacopterygian
osseans are categories invented by him. Linnasus adopted his icthyological
system unaltered in his Systema natura.
Linnaus's first great work: Systema naturae
This great work of Linnasus, the natural system "in which nature's three
kingdoms are presented divided into classes, orders, genera, and species,"
was published, as already mentioned, in Leyden in 1735. At the same time
was printed the above referred to Fundamenta botanka, and three years later
the important work Classes flantarum. These three really contain all that
is essential in the reform of classification which Linnasus carried out. Like
Ray, but in contrast to Tournefort, Linnasus as a systematician takes as his
starting-point the idea of species. He adopts Ray's theory of the species as
created from the very beginning and immutable, laying this down as a fun-
damental principle without limitations or exceptions. "We count as many
species as have been created from the beginning; the individual creatures
are reproduced from eggs, and each egg produces a progeny in all respects
like the parents." Thus there is no room for spontaneous generation, no
possibility for the seeds of one plant to give rise to a plant of a different
kind. Rather it was expressly maintained that in the beginning there was
created of each species one single pair, one of each sex, so that all individuals
of the same species possess a common origin. Again, there exist as many
SEVENTEENTH AND EIGHTEENTH CENTURIES XII
genera as there are, among the natural vegetable species, flowers (or fructi-
fications, as they are termed in these works) differing in number, shape, and
position. Classes are defined as a collection of genera that agree in regard to
fructification in certain main features. The order, again, is a subdivision of
the class which embraces a number of more easily summarized genera. The
application of these principles is Linnasus's universally known sexual system,
in which the classes are essentially determined according to the number of
stamens, and the orders according to the number of pistils. The practical
utility of the system is sufficiently evidenced by the fact that it is used to
this day in school education, although it was long ago abandoned in actual
scientific work. That this system, based as it was on only one organic system,
was one-sided, Linnaeus was very well aware, and in several instances he
departed from the fundamental principle merely in order to preserve the con-
nexion in certain groups which he found to be natural, as, for instance, the
classes Didynamia, Tetradynamia, and Gynandria, which, it is true, are
characterized by the stamens, but not only by their number, and which com-
prise forms that even the system of classification of our own day keeps to-
gether. For Linnaeus was fully aware that what should really be striven for
in botany was a "natural system": a classification of genera into groups
on account of a common similarity, not merely on account of the relations
of certain organs. He spent his whole life working out this natural system;
the results he achieved will be mentioned later on.
Linnasus's systematic classification of the animal kingdom cannot be
said to have turned out as successfully as his plant system. He divides animals
into six classes: (i) Quadrupedia, (2.) Aves, (3) Amphibia, (4) Pisces, (5) In-
secta, and (6) Vermes. Quadrupeds are characterized as follows: "the body
hairy, four legs, females producing live young, which they suckle"; birds:
"the body feathered, two legs, two wings, beak, females laying eggs" —
that is to say, purely external characteristics. Linnasus's precursor Ray based
his system, as we have seen, essentially upon anatomical characteristics: the
structure of the respiratory organs and of the heart; moreover, he differen-
tiated, although with faulty characterization, between vertebrates and in-
vertebrates; the latter he divides into four groups, while Linnaeus has only
two — all details in which Linnasus was undeniably inferior to his pre-
cursor. Fishes Linnasus has dealt with entirely in accordance with Artedi's
system, which he in fact acknowledges. Of the lower animals the only ones
that interest him are the insects. Moreover, Linnaeus has not laid down any
general principles for animal classification similar to his Fundamenta botanica.
Artedi's " Philosophia kthyologka" might certainly be said to have filled the
gap, but the latter's method, as we have already seen, differs not a little
from Linnasus's, primarily in the fact that his system is based on the genus
and not on the species — and in these circumstances it only remains to show
■LIZ THE HISTORY OF BIOLOGY
that the Linnasan reform was from the beginning more adapted to vegetables
than to animals.
The conception of species: their immutability
But, all the same, Linnjeus's contribution to the development of biology
has been of vital importance to science as a whole. In the first place, by fixing
the term ' ' species " as he did, he laid the foundation for the system of classi-
fication as it exists today. At the time when the dispute on the descent theory
was raging at its hottest, Linnasus, it is true, was exposed to the severest
censure just because he had declared the species to be immutable, as they
were created from the beginning — the dispute, in fact, raged just as much
over the belief in the creation as over the constancy of the species itself —
but in spite of the fact that the immutability theory is now abandoned,
the Linnasan species is used in practice by systematic science even today,
because it has not been possible to find any better substitute for it; a species
is regarded as the sum total of those individuals which resemble one another
as if they had a common origin. The other systematical categories which
Linnasus created also remain to this day, although some new ones have come
into existence as well. And quite as remarkable is Linnasus's influence on
what may be called the technical side of the classification system, which
he himself actually founded, exactly as it is applied today : his rules regarding
nomenclature, description, characterization, and synonymy have really
proved so complete that in principle posterity has had but little to add to
them. Moreover, the whole of this radical reform was carried out at one
stroke by a hitherto unknown young man after only a few short years of
utterly inadequate scientific training. This wonderful result was rendered
possible only by the fact that Linnasus combined exceptionally well-trained
powers of observation with an unparalleled natural genius for the formal
side of science. This latter gift was, so to speak, in his very blood: he had
a passion for classifying everything that came within his grasp; his medical
writings consist of groups of diseases in tabular form; his predecessors in
science he likewise classified under various headings, and once he even ar-
ranged, mostly as a joke, all his contemporary botanists according to mili-
tary rank, with himself as their general. For the fact that this mania for
classification never degenerated into mere dull pedantry he had to thank
his extraordinary love of nature and his passion and gift for observing life
in all its manifestations. It was this quality that prevented him from stag-
nating at the point to which he had so rapidly attained, instead of which
he spent his whole life striving to extend and perfect the science that he had
already so thoroughly recreated. These efforts, which, with the aid of his
pupils, he continued as long as his powers lasted, consisted in improving
the system he had already created, extending and perfecting the natural
vegetable system that he had already made it his ambition to work out,
SEVENTEENTH AND EIGHTEENTH CENTURIES ^13
and, finally, in making a number of observations of life in nature and the
interrelation of its different phenomena.
The binary nomenclature
The most important purely formal improvement of the system which Lin-
naeus effected was the binary nomenclature, introduced in 1753 into the
classification of the vegetable kingdom and somewhat later into that of the
animal kingdom. Previously he had followed the example of Tournefort in
characterizing every genus by a single term, while the character of the species
was designated by a short diagnosis of some few words. Now he introduced
instead one single character word for the species also, so that every plant
or animal received its character and its fixed place in the system by means
of only two words. This reform is certainly the most important of his con-
tributions in the purely formal sphere. Thanks to this alone biology has
been able to master the vast amount of form-material that has been col-
lected up to the present day, which could certainly never have been handled
if it had been necessary to employ diagnoses in order to denote the species.
His other reforms in connexion with the system applied not so much to
botany as to zoology; he left his botanical sexual system for the most
part undisturbed and contented himself with incorporating into it the new
species which were sent to him from all over the world. Of the improve-
ments which he introduced into the animal system the most worthy of
mention is the fact that he at last associated whales with the quadrupeds,
which resulted in the latter's receiving the name they have since borne —
Mammalia — that is, animals which feed their young. The orders into which
he divided this class, mainly after the dental structure, were, however, still
somewhat artificial and were long ago rearranged; on the other hand, his
method of associating man with the apes in the order Primates has been
retained. Birds, which were essentially classified according to their beaks,
have, as is well known, been still further regrouped. His transfer in the last
editions of Sy sterna natures of the Cartilaginei to the amphibians, which was
at variance with Artedi's system, was extremely unfortunate, based as it
was on a misconception of those fishes' gills. On the whole, Linnasus dis-
liked cold-blooded animals; as a motto for the amphibians he chose the
words: "Terrible are Thy works, O Lord," and he assures us that there
are not many who would wish to collect these animals. In regard to the inver-
tebrate animals he let his system stand unaltered; only the species, in par-
ticular the insects, were reduplicated like the plants.
The ' ' natural method of -plants
In spite of the enormous amount of work entailed in describing these new
animals and plants from all parts of the world, Linnasus found time to apply
himself to theoretical problems of great importance. Chief among these
should be mentioned his work on the natural vegetable system. As early
114 THE HISTORY OF BIOLOGY
as in the work Classes plantarum, published in Holland in 1738, he promul-
gated what he called "fragments of a natural method of arrangement": a
list of sixty-five "orders," each embracing a number of vegetable genera,
but without any characterization of the peculiarities that warranted their
being grouped together. By way of introduction he describes the natural
system as the highest, but hitherto unattained, aim of botany, which he
exhorts all truly distinguished botanists to strive after. For the creation
of such a system no particular parts of plants or flowers should, he maintains,
be used as a standard, but only the common agreement existing between all
parts of the plant. Several of the groups which he founded, such as palms,
grasses, Liliaceas, Umbellata, are still regarded as entirely natural. Through-
out the whole of the rest of his life Linnasus never let the natural system out
of sight, although he never thought that he would complete it. In his Philo-
sophia botanica(i-j'^i)hc again cites a number of natural groups, now provided
with names, and in doing so points out that the vegetable groups everywhere
border on one another, like the countries on a map of the world. In point of
fact, his realization of the difficulty of trying in a comprehensible way to
present the natural affinities of living creatures was a proof of his keen eye
for the infinite multiplicity of nature; his caution might well be borne in
mind by many a biologist of our own time who has rashly drawn up a genea-
logical tree for some animal group or other. In connexion with this feeling
of Linnxus for the difficulty of determining natural affinity, it is worth
mentioning that in his later writings he discusses with far greater caution
than in his earlier years the question of the bordering of the species on one
another. It was not only that he had seen masses of varieties overlapping one
another, but he had also observed the altered forms produced by hybridiz-
ing — he himself was very successful in hybridizing in his own garden —
and as a result of all this the delimitations of species, which he once felt to
be so certain, began to be obliterated. The doctrine of the original creation
he certainly could not abandon, but he began to consider the possibility of
the genera's having been created and only one or a few species of each, and
afterwards new species' being able to arise out of the old. In the final edition
of Systema natura he has omitted the definite assertion that no new species
arise. He who has so often been accused of dogmatism was really less dog-
matic than many modern scientists who have proved themselves ready to
accept blindly the prevailing theories of the day.
In the above-mentioned Philosophia botanica Linnasus has also expounded
an organic theory in respect of the vegetable kingdom. A great many of his
clearly formulated characters of the various parts of plants are still valid to-
day. Many consider the anatomical section of this work to be weak, even as
compared with the investigations of the earlier botanical anatomists Mal-
pighi and Grew. This may be true, for Linnasus was, generally speaking, no
SEVENTEENTH AND EIGHTEENTH CENTURIES 115
anatomist; true, he did not fail to urge the study of anatomy as well, but his
own gifts lay far rather in work upon living nature than at the dissecting-
table. Of his purely morphological observations, on the other hand, many
are of lasting value; he thus established the fact that all leaves, both
plant leaves and flower-petals, go through a common process of develop-
ment, a discovery very often attributed to Goethe. But when he tries
his hand at comparative anatomy, he usually fails, as when he com-
pares the parts of plants and animals: marrow and spinal marrow, skin and
bark, etc.
Phenologkal and geographical biology
On the other hand, Linnreus's contributions to the knowledge of the con-
ditions under which plants and animals live in their natural state are excep-
tionally many-sided. These natural observations of his, which occur scattered
throughout his disputations and platform speeches, bear witness not only to
his keenness of observation, but still more to his ability to combine and draw
conclusions from what he observed. Thus in the course of a graduation speech
' ' On the Rise of the Habitable Earth, ' ' which begins with a discussion of how
all vegetable species were able to grow at once in paradise — they must
have existed there, for otherwise Adam would not have been able, as stated
in the Bible, to give them names — he expounded a theory of the propaga-
tion of plants, based on universal research-material, which was so well ar-
ranged that it still has its value at the present day. In other dissertations he
has contributed to the knowledge of the "stations" of plants (nowadays
termed "locations") and has described the influence of external conditions
upon the size, florescence, and distribution. All that is now called pheno-
logical, ecological, and geographical zoology and botany has consequently
its origin in him. Finally, in the disputations "Politia natura" and " CEconomia
naturx" he gives a radical explanation of all that we moderns call harmony
in nature: that all living creatures are adapted to certain conditions of life
and that the various plants and animals through their activities keep nature
in equipoise, so that "every vegetable species has been given its special insect
for the purpose of keeping her under control and to prevent her from spread-
ing too much and ousting her neighbours," while the Hymenoptera Para-
sitica and small birds look after the insects, and birds of prey after the small
birds. That he lets all this take place under God's constant guidance, to His
honour and for the benefit of man, should not in our time detract from
the value of the observations and the wealth of ideas expressed in these
works.
The balance which Linnaeus thus found in nature he sought also in the
ethical sphere through his well-known speculations upon the "Nemesis
divina," which, however childish they may be in their detail, are neverthe-
less typical both of the man himself and of his time; both Leibniz and Vol-
Xl6 THE HISTORY OF BIOLOGY
taire^ likewise ruminated over the righteousness governing the cosmic
process. And in this, too, Linnjeus perceived the divine guidance that he so
earnestly sought in natural phenomena. He was an optimist all through —
one of the few happy beings who could see harmony everywhere because they
have had such a harmonious disposition themselves. He regarded his life's
work with a mixture of naive self-satisfaction and humble gratitude to the
Almighty, under whose guidance he was always conscious of living. And he
may well have been satisfied, for in the science he so faithfully served, few
have exercised so great an influence as he.
Pupils of Linnceus
It has been pointed out above that Linnaeus possessed an extraordinary
power of gathering pupils round him and interesting them in facts and ideas
in the science he represented. Naturally they were for the most part Swedes,
but a number of foreigners also came to hear him. His Swedish pupils, after
receiving their education, were generally sent to various foreign countries
in order to make collections and to describe the places they visited. Linnasus
had drawn up for them special instructions, which might serve equally well
today as a guide for research-workers in a foreign country. These pupils were
travellers and collectors; as a general rule, they made no independent dis-
coveries. Several of them fell the victims of hardship and disease, some ob-
tained distinguished appointments abroad, and others returned home. As
examples of these collectors may be mentioned F. Hasselqvist, who travelled
for three years in the East and died in Smyrna in 1751, and P. Lofling, whom
Linnasus called his most beloved pupil and who, on the invitation of the
Spanish Government, worked first on the Pyrenean peninsula and then in
South America, where he died in 1757. Further, Per Kalm (1716-79), the
first biologist in Finland to work independently, who in the course of a
three years' sojourn in North America made valuable contributions to that
country's natural history and national economy, and who afterwards acted
as professor of economics at Abo, strove incessantly by the application of
natural science to practical life to further the material development of his
country. Kalm's fellow-countryman Peter Forskal (i73x-63) first studied
2 Voltaire, it will be remembered, after the terrible earthquake that took place in Lisbon
in 1755, wrote a poem in which he asks Providence of what those innocent men were guilty
who perished in it. Linnaeus takes up the same question; he consoles himself with the considera-
tion of the many innocent people whom the Inquisition had burnt at the stake in that city and
recalls that the earthquake took place on All Saints' Day — the same day on which the autos-
da-fe used to be held. A genuine Old Testament-like idea; the city had sinned and was punished
accordingly. Undeniably more attractive are some of the innumerable examples of Nemesis
which Linnaeus cites from private life, such as the story of the lady who struck her servant for
having fallen downstairs and dropped a precious china bowl; the same evening the lady herself
fell down the same stairs and broke her leg. Generally speaking, Linnxus possessed a strongly
democratic turn of mind, a lively sympathy for the poor and oppressed.
SEVENTEENTH AND EIGHTEENTH CENTURIES xiy
natural science at Upsala, then philosophy at Gottingen; owing to an essay-
he wrote attacking the Wolffian philosophy, which was predominant at
that time, he was unable to obtain an appointment in Sweden, with the
result that he went into Danish service as natural scientist on an expedition
to the East. There he died, leaving behind him singularly valuable collec-
tions. Another who took a post abroad was Daniel Solander, who had an
appointment in the British Museum, London, and died there in lySz; and
finally may be mentioned Karl Peter Thunberg (1743-1818), who held
Linnasus's professorship from 1784, after having travelled for nine years in
eastern Asia, particularly in the then unknown country of Japan, and col-
lected a rare amount of material in the way of plants and animals. More
independent than any of these others, however, was the Dane, Johan
Christian Fabricius (1745-180S). The son of a physician, he became an un-
dergraduate in Copenhagen in 1761 and afterwards spent two years at Upsala
with Linnasus, with whom he formed a lifelong friendship. After returning
home he published several valuable works on entomology, in which he
applied Linnasus's method to the insects — the titles of his works corre-
spond to Linnasus's, as will be seen from the bibliography at the end of this
book — and greatly increased the knowledge of that class of animals. Abroad
he was highly esteemed; at home, on the other hand, he received but little
encouragement. After a long period of waiting he eventually became professor
at Kiel, on very poor terms, wherefore he spent most of his later years abroad,
chiefly in Paris, where he had many friends. Of Linnasus's personal pupils
he was perhaps the one who, besides strictly applying his master's system,
likewise understood best how to employ it for his own researches, which
were of lasting value.
Development of systematic biology after Linnaus
For it was not long before the Linnasan natural science began in a general
way to degenerate into a spiritless task for collectors and describers, who
merely aimed at discovering and incorporating in the system as many fresh
species as possible, at the very highest in the hope of being able to use them
to some practical purpose for the benefit of mankind — an idea which very
much attracted that "age of utility" and which, it is true, Linnaeus himself
also strongly emphasized. On the other hand, they neglected to cherish and
develop those ideas for the future by which Linnasus himself set such store —
the natural system and the study of the conditions of life in nature. The
result was that the system of descriptive classification, which has so often
been called Linnasan science, actually became an expression for a quite
limited part of the master's high aims; it certainly became, and has remained
so to this very day, a necessary basis for the future progress of biology, as
it is also a pedagogically indispensable introduction to that science; but it
has also been possible to practise it without any deep insight into the phe-
Il8 THE HISTORY OF BIOLOGY
nomena of life and has thereby not infrequently acquired the character of
merely a collector's hobby.
In fact, this seems to have been the fate which threatened biology as a
whole during the era at present under discussion; that it did not actually do
so is due to a large extent to the fact that contemporaneously with Linn^us
there was another scientist at work who was aiming at leading science into
a direction entirely different from his, and who, at any rate partially, suc-
ceeded in his efforts. This man was Buffon.
CHAPTER VIII
BUFFON
His Studies and career
GEORGES Louis Leclerc de Buffon was born in 1707 at Montbard, in
Burgundy. His father was councillor of the Burgundian parlement
• at Dijon, the capital of the province, and thus belonged to that
bureaucratic nobility which was so influential in France in earlier times and
which gave to the country many of its finest men, who were often remark-
able for their cultural interests and their wealth. Both existed indeed in the
home in which young Buffon was brought up; he received a thorough educa-
tion in his native city and had good prospects in the career which his family
had long followed, when chance turned his footsteps into a different direc-
tion. He' made the acquaintance of a young Englishman, Lord Kingston,
who was travelling on the Continent accompanied by a tutor who had
studied natural science, and travelled with him through France and Italy.
During this tour Buffon's interest in nature, which was to be the dominating
factor in his life, ripened. He accompanied his friend to England and spent
a year in London studying, particularly mathematics, physics, and botany —
sciences which were at the height of their development in the country that
gave birth to Newton and Ray. Having returned to France, Buffon published
a translation of Newton's fluxions, as well as of the English botanist Hales's
Vegetable Staticks — two works which presaged the direction that his own
activities were to take. As he was wealthy, he was able to devote himself to
regular scientific labour, first of all directing his attention to mathematics
and physics. In 1739 ^^ '^^^ elected an associate of the French Academy of
Sciences and in the same year was appointed "keeper of the Jardin du Roi,"
a post of some distinction, which was still further enhanced by his activities
that resulted in the Jardin du Roi, now the Jardin des Plantes, becoming the
centre of biological research in France. In the period that followed, his great
gifts proved of benefit both to himself and to the science he had chosen: he
succeeded in arousing a general interest for natural science amongst the lead-
ing circles in France, so that even the King, Louis XV, who was so indifferent
and such a stranger to all ideal interests, granted large sums for the improve-
ment of his garden and to assist the scientific work carried out there, while
many other eminent personages likewise patronized this science and its dis-
tinguished representative. Buffon himself was created a count, was made a
119
XIO THE HISTORY OF BIOLOGY
member of the French Academy, and was in many other ways honoured by
the great. It was also his fortune to play a brilliant part at a time when bril-
liant qualities were valued more than usual. He was of handsome person and
stately presence, which was further enhanced by the exquisite care devoted to
his dress and outward appearance; he was an excellent stylist and orator,
and, although comparatively reticent in society, he knew how to entertain
in a manner befitting his social position. Naturally he also had his enemies in
the scientific world, who caused him much annoyance both in public and in
private. Even the theological faculty in Paris was not satisfied with him, as
his views did not seem to be entirely orthodox, and there was once a ques-
tion of arraigning him. The distinguished man of the world, however, who
naturally had not the least inclination to become a martyr, parried the ac-
cusation with a few elegant courtesies about the infallible authority of the
Church, and so the matter was allowed to drop. At the same time, in pri-
vate letters to trustworthy friends he expressed extremely sceptical opin-
ions, which place him in utter contrast to Linnasus, with his childishly
naive piety. These two were destined to become antagonists in other spheres
also. — Active to the last, BufFon attained an age of over eighty, dying in
1788. His only son, whom he desired, but in vain, to become his successor,
died by the guillotine during the Revolution.
His Histoire naturelle
At quite an early age BufFon had believed it to be his mission in life to write
a general natural history, an account of all the knowledge of nature that
could be amassed, and in 1749 the first part of his Histoire nafurelle was pub-
lished — a work upon which he was engaged for the rest of his life. In the
preparation of this work he associated himself with the eminent anatomist
Louis Daubenton (1716-1800), who was for a long time conservator under
him and afterwards became a professor at the College de France. He carried
out the anatomical and morphological detail-work, while BufFon had the
management of the whole and was responsible both for the ideas incorpo-
rated in it and for the method of presentation. The original edition comprised
fifteen volumes, the first of which dealt with general natural science, and the
remainder with man, the mammalians, and the birds. BufFon got no further
in the sphere of biology. In some supplementary volumes, which came out
later, a number of general natural-scientific questions were discussed, includ-
ing mineralogy, a science on which BufFon was weakest, as he was no chem-
ist. Later on, several editions of the great work were published — a proof of
how popular it was, in spite of its expense. The main reason for this was
undoubtedly BufFon's brilliant style. Not only did he produce vivid descrip-
tions of the nature and habits of animals, the like of which had never been
read before and seldom have been equalled since, but he also succeeded in
dealing with the most difficult physical and cosmological problems in a
SEVENTEENTH AND EIGHTEENTH CENTURIES XZI
clear and easily comprehensive style. Moreover, besides these external quali-
ties, his work possesses immense practical advantages: it contains the idea
of a new and magnificent conception of natural science, and particularly of
biology, and its influence on the future development of the latter science has
been exceedingly great.
BufTon began as a physicist; as we have already seen, he translated a
work of Newton's, and he also studied Leibniz; he had at once been struck
by the wonderful obedience to law that, according to the then new physical
and astronomical discoveries, governs the universe. In the light of these
discoveries the cosmos appears as a mighty piece of mechanism, which works
according to given laws, and in which both the past and the future can be
mathematically calculated. Is it not likely that in such circumstances the
phenomena here on earth, both in inanimate and animate nature, would also
be subject to a similar obedience to law? That is the question Buffon has put;
he has answered it in the affirmative and he has tried to give proofs of it.
His lasting service to science lies in the fact that he thus endeavoured to
incorporate biological phenomena in their entirety as a link in the great law-
bound world-process; thereby he made a great advance towards the goal that
our modern natural science has set itself, and progressed far beyond the mech-
anistic biologists of the seventeenth century, the Borellis, the Perraults, and
others who only sought to apply the laws of mechanics to the human body,
without any more universal objects in view. That Buffon, with the limited
material of facts available, could not succeed in creating a theory capable of
passing the test of modern knowledge is quite obvious, but that does not
prevent us from acknowledging the greatness inherent in his very ideas, and
the ingenuity with which he attempted to carry them out.
His general vieivs
Buffon introduces his natural history with an account of the general princi-
ples on which he considers such a history should be written. Here he at once
expresses his view of nature as one whole, all of whose forces gear into one
another and all of whose manifestations stand in mutual causal connexion.
But at the same time he utters a warning, in words reminiscent of Bacon,
against bringing the multiplicity of nature under too simple points of view;
with an obvious allusion to Linnasus he warns us against those who speak,
for instance, of a mineral growing, and who compare in detail the organs of
animals with those of plants; it is, he says, trying to compel nature to come
under our arbitrary laws, not ascribing to the Creator more ideas than we
ourselves possess. The vast wealth of nature must rather be realized and ac-
knowledged from the beginning; the first causes of its phenomena will always
be hidden from us, and what remains to be done is to observe a number of
particular phenomena, compare them, and in them try to find a regular course
of events. It is thus impossible to create any universal system covering all
±1.2. THE HISTORY OF BIOLOGY
natural phenomena; its forms and manifestations imperceptibly merge into
one another, wherefore vegetable systems in particular, such as those set up
by Tournefort and Linnaeus, are utterly unnatural. Buffon is very keenly op-
posed to Linnasus; he asks ironically what is the use of the sexual system
when the plants have ceased to flower. In fact, the whole of the Linnasan
system of species classification was intrinsically repugnant to Buffon; it
seemed to him to be an arbitrary decimation of unified nature into little bits;
Linnasus's efforts to create a natural system and his emphasis upon the in-
completeness of the classification system, in which he did not vary very
much from Buffon's own ideas, were simply neglected by him; apparently
they appeared to him merely as slight inconsistencies in a falsely founded
view. He criticizes Linn^eus's animal classification with similar asperity and
undeniably touches its weakest point when he rejects the two great classes,
insects and worms; no one, he says, can imagine that crayfish are insects,
and shells worms. Instead of six Linnasus should have set up twelve classes,
or even still more, for the more groups there are the nearer we arrive at the
truth. In fact, in nature there are only individuals; genera, orders, and classes
exist only in our imagination. In this Buffon is undoubtedly right in theory,
but he overlooks the practical advantage of the "imagined" categories,
without which the various life -forms could not possibly be dealt with by
science. Instead of the artificial classification-system which he thus rejects,
Buffon presents an introduction to the study of nature that, to some extent,
is reminiscent of the modern intuitive method of instruction; the description
of nature should follow the course which a man ought to pursue if, after
having forgotten all that he ever knew, he were put in a place surrounded
by natural objects; he would first learn to differentiate between animals,
plants, and stones, and then, as regards animals, he would observe the most
essential features in their habitat and mode of life and would group the in-
dividual animals accordingly in his mind and finally would learn to compare
the different animals with one another in greater detail, first distinguishing
the tame animals from the wild, then among the wild those who lead the
same mode of life and resemble one another in their structure. He glorifies
the ancient biologists Aristotle and Pliny, just because they followed a
similar natural plan of dealing with living creatures. In his opinion, how-
ever, modern research should in no way confine itself merely to observing
and describing; the scientist should rather confirm his observations by means
of experiment; he should know how to combine observations and to gen-
eralize facts, to make individual phenomena obedient to general laws, and,
finally, to compare the most comprehensive phenomena of nature with one
another. The ultimate aim is to bring all phenomena under the general laws
of physics, those laws whose causes remain incomprehensible to man, while
only their effects are perceptible. Here Buffon has undoubtedly learnt from
SEVENTEENTH AND EIGHTEENTH CENTURIES ^2.3
Newton and, either through him or directly also, from Galileo; at any rate,
he here displays an insight into both the aims and the limitations of natural
science that few scientists before his age possessed and that many have lacked
even in our own time.
His history of the earth
BuFFON, having thus given an account of the general principles on which
naturalists should work, proceeds to give a theory of the earth and its de-
velopment into a habitation for living creatures. This problem was mani-
festly of very great interest to him; in fact he has dealt with it repeatedly —
in an essay at the beginning of this great work, called "Theorie de la terre,"
and in another, more extensive essay, written considerably later, entitled
"Des epoques de la nature." In this sphere, it is true, he had precursors; Steno,
whose geological works have been mentioned previously, Ray, who wrote
a treatise on the changes in the earth, and, the greatest genius of them all,
Swedenborg;^ but Buffon must nevertheless be mentioned as the first who
thought to investigate the earth's history with special reference to the
development of living creatures.
Moreover, in his latter work he had the temerity to reject the biblical
six-thousand-year age of the world and to attribute to the earth a far higher
age — certainly small in comparison with the length of the epochs which
modern geology assumes, but at least entailing a break with the till then
incontrovertible theory of the creation — a break induced by the impossi-
bility of fitting the geological and biological evolution on the earth within
such a short space of time as six thousand years. Steno had already faced this
dilemma, but, pious Catholic as he was, he preferred to abandon geology
rather than Church doctrine; BufFon courageously took the step in spite of
his previous contretemps with the French theologians. Even as early as in his
"Theorie de la terre" he expounds his ideas as to the origin of the earth,
expressly emphasizing, however, their purely hypothetical character. He as-
sumes, in agreement with Leibniz, that the earth evolved from an incandes-
cent state, but while the latter believed that the earth itself had from the
beginning been a "sun," Buffon derives it from the sun's mass, assuming that
once upon a time a comet collided with the sun, with the result that pieces
broke off which gave rise to the earth and the other planets. This hypothesis
has often been cited as a proof of Buffon' s extravagant imagination; really,
it is no more eccentric than many contemporary cosmogonies, in which
comets in general quite often played a part, and in fact Buffon's is presented
with much more reservation than the others. After the incandescent state
there followed a period when the seas covered the earth, when the tide
exercised great influence upon earth-formation. As a proof of this theory of
^ Buffon is said to have known Swedenborg's cosmological theories; he never quotes
them, however, though he does quote Steno and Ray and some other, less important authors.
Z2.4 THE HISTORY OF BIOLOGY
the seas' wide distribution BufTon cites the discovery of fossilized marine
animals, especially shells, up in the mountains, and even the stratified
nature of the geological formation in general. ^ In the essay "Epoques de la
nature" BufFon divides the history of the earth into seven periods: (i) when
the earth and the planets were formed, (x) when the great mountain-ranges
were created, (3) when water covered the mainland, (4) when the water
subsided and the volcanoes began their activity, (5) when elephants and
other tropical animals inhabited the North, (6) when the continents were
separated from one another, (7) when man appeared. It would take too
long to give a more detailed account of his description of these periods. In
this geological theory he makes Vulcanism in general play a more important
part than in the earlier work, and the tide becomes of less significance. The
greatest service he rendered, however, is that he clearly realized the change
in the animal and vegetable kingdoms from epoch to epoch; he tried to work
out a "natural history of creation," based on law-bound evolution; he specu-
lates upon the origin of the various life-forms and the place of their appear-
ance and combines these two circumstances with calculations as to climatic
changes and other purely physical conditions — in all this a pioneer of the
conception of nature which was not generally accepted until a century after
his time.
His biological theory
BuFFON has expounded his biological theories in a volume bearing the title
Historic naturelle des animaux. It begins with an investigation into the differ-
ence between animals, vegetables, and minerals and establishes the fact that
there is no absolutely definite boundary between the animal and the vegetable
kingdoms, but that transition forms may exist; common to both kingdoms
is the individuals' power of giving rise by means of reproduction to new
individuals like themselves. Another common property is the power of
growth; this shows that a fundamental agreement prevails amongst all liv-
ing creatures, in spite of differences in detail, whereas only matter as a
fundamental substance is common to animate and inanimate things. Both
animals and vegetables arise as species, the criterion of which is that they
propagate; on the other hand, there is no question of a common creative
origin. On the whole BufFon refuses to see in the origin of life the result of a
^ As an example of the lengths to which it was still possible to go during the " ' enlightened
eighteenth century in explanation of natural phenomena, it may be mentioned that Voltaire,
who, however, would pass as a disciple of Newton's, declared that Buffon's theory of the origin
of fossils up in the mountains was irrational; the shells that had been found there had probably
been left there by pilgrims who took them from the East. The two geniuses consequently came
into serious disagreement; but later they were reconciled and Voltaire declared BufFon to be
a second Archimedes. Buffon capped this compliment with the assurance that no one would
ever be called Voltaire the Second. Voltaire's flattery shows, however, that BufFon was con-
sidered, and indeed wished to be considered, primarily a physicist.
SEVENTEENTH AND EIGHTEENTH CENTURIES 1x5
particular act of creation; life, says he, is not a metaphysical characteristic
of living creatures, but a physical quality of matter.
Since, then, the most vital quality of life is the power of reproduction,
Buffon devotes very close study to it. For this purpose he does not start from
the most highly organized creatures, but begins his investigation — this,
too, a modern feature — with the most primitive form of reproduction —
that by means of division in plants and primitive animals. Why is it that a
severed branch of a twig of a tree grows up into a new tree, that a piece of
polypus gives rise to a new polypus? Buffon answers this question by assum-
ing that the plant and the animal are composed of a mass of particles formed
like the individual in its entirety, and which therefore, when they become
detached, can develop further and form a new individual of the same kind.
This theory of independent particles, the idea for which he undoubtedly got
from Leibniz's monad theory, Buffon further develops to form the basis of
his conception of all the phenomena and functions of life; just as inanimate
matter is composed of an incalculable mass of minute particles, so there
exists in nature a vast number of organic particles that are animate and
formed like animate beings. ' 'Just as there may be required perhaps a million
minute salt cubes to form one grain of sea-salt, so it would take millions of
organic particles similar to the whole to form a bud containing the individual
of a tree or a polypus." By making this assumption Buffon also seeks to get
rid of the preformation theory, which was generally embraced by his con-
temporaries and which he keenly criticizes, maintaining among other things
that it would presuppose an infinite number of daughter individuals con-
tained in the original mother animal, which in itself is an entirely irrational
supposition. But when it comes to setting up an acceptable theory of sexual
reproduction in place of the preformation theory, Buffon comes to realize,
as he himself openly acknowledges, that it is easier to destroy than to build
up. He founded a general physiological hypothesis according to which
animals through the food absorb a quantity of these ubiquitous organic
particles and in the various organs of the body assimilate from these what the
body requires; whatever is left is collected in the genital organs and gives
rise to individuals like the parents. That the embryo is thus formed by a
combination of a mass of minute independently living particles he believes
to be proved by the spermatozoa existing in the seminal fluid; that the female
sexual product actually consists of similar minute beings he also believes he
had proved by a microscopical study of the ripe follicles in the mammalian
ovary; in their fluid he believed that he had found mobile life-elements
similar to those in the semen which he actually illustrates^ and which in his
^ What Buffon and his collaborators — he quotes several, including the English micro-
scopist Needham — actually saw in the follicular fluid it is difficult to say; perhaps detached
cells from the follicular epithelium; maybe also coagulation products.
il6 THE HISTORY OF BIOLOGY
view combine with the spermatozoa to form the new individual. In connexion
with the question of evolution, reproduction, and growth, BufFon sets up a
very abstract and difficult hypothesis concerning the mutual connexion of the
different parts within the organism; he believes that each individual repre-
sents a "moule interieur," by which he apparently means the constant form of
every living creature, attained with the co-operation of the organs, which
are fed and grow by assimilating these living particles that fill the whole
of nature and are the one essential in the assimilation of food — of both
animals and vegetables — in growth, and in reproduction. This theory of
living particles thus forms the very corner-stone of Buffon's biological
speculation and is both its strength and its weakness; with its aid he avoids
the difficulty of explaining the origin of life without the assumption of a
supernatural act of creation — he does not expressly deny such an act, it is
true (that would have been too daring for his age), but it is quite obvious
that he will have nothing to do with it — on the other hand, he had to make
good with assumptions which very much resemble the ancient spontaneous-
generation hypotheses, which had already been rejected by the biologists of
the seventeenth century. At any rate, of greater value than the results of these
speculations is his criticism of the actual method of natural research, which
even in modern times makes profitable reading; his own theories he in no
wise propounds as if they were proved truths, and his warnings against con-
fusing hypotheses and facts many a modern biologist might well take seri-
ously to heart.
Besides these purely biological questions Buffon also discusses psy-
chological problems. His speculations on animal psychology are, however,
of little importance; he certainly admits the existence of intelligence in
animals, in contrast to Descartes, but he denies that they possess memory
and reflection. On the other hand he has some striking observations to make
on the domestic animals' intellectual dependence upon human training, as
well as on their sense-impressions and the varying power of the latter.
On man
Like Linn^us, Buffon treats man as a natural-history subject and gives a
detailed description of man's evolutional history, alimentary conditions,
and habits of life, which has justly become famous, not merely for its impor-
tant formal merits, but also as being the first attempt at anthropology in the
modern sense. Though human anatomy had already been thoroughly dealt
with in the sixteenth and seventeenth centuries, nevertheless a universal
treatment of man in regard to his entire relation to nature was something
quite new. From an anatomical and physiological point of view he certainly
has not very much that is fresh to relate, but he conscientiously and critically
summarizes the existing scientific material; he gives an account of the devel-
opment of man from embryonic life through the various ages; he tries to
SEVENTEENTH AND EIGHTEENTH CENTURIES 2.Z7
analyse the growth of speech in the child, and the influence of mental emo-
tions upon facial expression; he discusses the power of mental perceptions to
reproduce reality and insists on science's dependence upon them. He investi-
gates the circumstances of death at various ages; he gives statistics showing
the mortality in certain French provinces and seeks with the aid of "proba-
bility calculations" to ascertain the longevity of various ages; he compiles
data with regard to the peculiarities of wild tribes and abnormal phenomena
in civilized peoples. Above all, he maintains that man has his bodily func-
tions in common with the animals, but that, on the other hand, there is a
fundamental psychical difference between them, which renders it impossible
to compare human and animal intellectual qualities.
His Xoogra-phy
Of the animals Buffon, as already mentioned, had time to complete only the
quadrupeds and birds, which are dealt with in detailed monographs covering
each species. Naturally, these are each of varying value; all of them however
share in common a brilliant exposition and a universal treatment of the sub-
ject, in striking contrast both to the earlier zoographers' motley mass of
notes and to Linnasus's brief, summary diagnoses. As a describer of nature
Buffon is of fundamental importance, in certain respects still unexcelled. It
would take too long to enter into the peculiarities of his zoography; it need
only be pointed out that it is not merely formal services that have earned it
its well-merited fame, but in many of his descriptions, particularly in those
of birds, there are a number of keen and striking detailed observations as to
mode of life, reproduction, and other biologically interesting factors.
Daubenton s cofnparative anatomy
To each monograph on mammalian animals Buffon 's collaborator, Dauben-
ton, has added an account of the animal's anatomy. By way of introduction
he sets forth the principles on which to base this kind of general — nowa-
days we should say "comparative" — anatomy, in contrast to the descrip-
tive, which had hitherto been practised. He considers that all animals should
be investigated in respect of their most vital organs — bone-structure, heart,
brain, respiratory, digestive, excretive, and sexual organs — and the results
thus obtained compared. Following this principle, an account is given of
every mammal's anatomy, particularly the bone-structure: the bone-struc-
ture of the horse is compared in detail with that of man, and the bones of
other animal species are mutually compared. Such a comparative examination
of the anatomy of various animals carried out on a uniform plan was at that
period something new and proved of great significance for the future; the
part played by comparative anatomy in modern biology is too well known to
need any special emphasis here.
Thus Buffon carried out his plan of presenting nature, both inanimate
and animate, as one whole, evolved and held together by purely mechanical
1X8 THE HISTORY OF BIOLOGY
laws. In this, however, he was not entirely consistent. In a curious treatise
entitled Homo duplex he has described man as composed of two principles,
fundamentally distinct from one another, one spiritual and the other material,
of which the material develops first and predominates in the embryonic stage
and during childhood, while the spiritual appears later and is developed by
means of education and training, without which it would lead to stupidity
and vain delusions. This dualistic conception of man might well appear to be
fully in accordance with the principles that had till then been and were still
at that time officially recognized — it might be suspected that BufFon here
made a concession to the ecclesiastical authorities who had persecuted him —
had not the style of the whole been so utterly different from all that is under-
stood by conventional religion. Instead we here come across a trait in Buffon
which we should not expect to find in that exceedingly brilliant and suc-
cessful man — namely, a deep pessimism. To his mind, the contrast between
spiritual and material appears most marked in those attacks of melancholy
and listlessness when one lacks all power of decision, when one "does what
one would not and would do what one does not ' ' : when one feels that the
personality is divided into two, of which the one part, reason, indicts the
other without being able to overcome its resistance; sometimes reason wins,
and then one performs one's duties gladly; sometimes the flesh wins, and
then one indulges in pleasure, but sooner or later these unhappy hours and
days return when disharmony prevails. Especially vivid is the passage in
which Buffon describes how love, which makes animals happy, simply
makes men wretched; in words of wild despair he depicts the vainness and
folly of this passion, which certainly brings with it bodily satisfaction, but
is morally valueless and only calls forth jealousy and other degenerate feel-
ings. This melancholy conception of life was, as a matter of fact, in no way
peculiar to Buffon; it was, on the contrary, as we shall see later on, a wide-
spread view during the epoch to which he belonged.
Buffon' s influence
Buffon has played a fundamental part in the history of biology, not on ac-
count of the discoveries he made, but on account of the new ideas he produced.
Those ideas that he brought out, which he was able only imperfectly to
realize in detail, have since then been taken up by others, who, having better
opportunities for obtaining actual scientific material, have applied them in a
wider sense: thus, Cuvier, the pioneer of comparative anatomy and paleon-
tology, adopted many of Buffon's fundamental ideas; similarly Bichat, the
originator of the tissue theory, in his sphere, as well as Lamarck, with his
theories of the evolution of living organisms, in that field, has manifestly
felt the influence of Buffon's speculations. Through these scientists many of
the ideas produced by Buffon have now been incorporated in the general
knowledge of natural science. If in spite of this he has often been depicted,
SEVENTEENTH AND EIGHTEENTH CENTURIES ^19
especially beyond the borders of France, as a talented dilettante, a witty
popular scientist and writer, this is mainly due to the relations in which he
stood, both in his lifetime and long afterwards, to the representatives of the
Linnnsan system of classification; these latter, who for a long time felt that
they were the sole upholders of a truly exact natural science, looked compas-
sionately down upon BufTon's unsystematic descriptions and imaginative
speculations. When, then, the dominion of Linnasanism fell, the comparative
and speculative lines of research which succeeded it already possessed entirely
different intellectual material to build upon, and Buffon's theories thereafter
necessarily appeared vague and childish. His services, however, must in all
fairness be duly acknowledged; in the purely theoretical sphere he was the
foremost biologist of the eighteenth century, the one who possessed the
greatest wealth of ideas, of real benefit to subsequent ages and exerting an
influence stretching far into the future.
CHAPTER IX
INVERTEBRATE RESEARCH IN THE EIGHTEENTH CENTURY
Successors of the great seventeenth-century biologists
THE EIGHTEENTH CENTURY displays OH the whole great activity in the
sphere of the natural sciences. In physics and chemistry the successors
of Newton and Stahl worked at extending the spheres to which
their masters had opened the way. In the realm of biology Linnasus and his
disciples held sway and were fully occupied in incorporating known species
into the system and in discovering fresh ones. As already mentioned, Buffon's
activities belonged rather to the future. But besides this there worked during
the eighteenth century a number of naturalists whose achievements connected
them more or less directly with the great biologists of the preceding century.
Towards these pioneer anatomists, microscopists, and physiologists their
successors during the eighteenth century have to some extent the character of
Epigoni: they made no such epoch-making discoveries as Harvey's or Mal-
pighi's, but, on the other hand, they took advantage in many and various
ways of the discoveries that had already been made; the problems which
had already been of direct importance were discussed from various points of
view, while at the same time ideas were expressed in more than one field of
research which gave a presage of future ends to be gained. Malpighi's and
Swammerdam's investigations into the anatomy of the lower animals were
thus resumed and carried a step further, as also Borelli's and his successors'
physiological work; Leeuwenhoek's and de Graaf's discoveries in the field
of reproduction were elaborated, and the discussion between the animalcul-
ists and the ovists went on, particularly in the first half of the eighteenth
century, with undiminished liveliness; the epigenesis and preformation theo-
ries were also keenly debated, although at first with the balance decidedly
in favour of the supporters of preformation. Later in the century, however,
contributions were made in these very spheres which put a different complex-
ion on those questions. And even for several other spheres of biology the
latter half of the eighteenth century represents a period of decisive prepara-
tion for the development that took place during the nineteenth century. —
In the following paragraphs we shall review, to begin with, some of the
more important contributions to the biology of the lower animals, and
afterwards the advances made in the spheres of anatomy, physiology, and
evolution.
130
SEVENTEENTH AND EIGHTEENTH CENTURIES 131
Rene Antoine Ferchault de Reaumur was born of noble and wealthy
parents in the year 1683. He received his education at a Jesuit college, after-
wards studying jurisprudence in Paris, but he soon abandoned that career and
applied himself whole-heartedly to natural science. Having inherited a for-
tune, he was able to lead the life of a private scholar; membership in the
French Academy of Science was the only distinction he obtained. He died in
1757-
Reaumur was active in many branches of natural science, both theoreti-
cal and applied. He invented improved methods of iron-refining and made
important contributions to our knowledge of the expansion of gases and
fluids and of specific heat. He is best known for his invention of the eighty-
degree thermometer scale, which bears his name and which is still used in
many countries. In biology, too, his activities have been many-sided and
important. His greatest and most famous work is his Memoires pour servir a
rhistoire des insectes, a work in six large quarto volumes. This work is un-
doubtedly of fundamental importance in insect biology and is in fact one of
the most monumental works written in this field of research. It offers a
number of extremely valuable contributions to the knowledge of the anatomi-
cal structure of the insects, their evolutional history and conditions of life.
His chief master is Swammerdam, whose system he in the main adopts, but
he considerably widens the sphere of the latter's researches. True, he did
not possess the master's incomparable ability in the work of preparing mater-
ial, but instead he had at his disposal a greater wealth of material for his
researches, while a long life made it possible for him to carry out lengthy
and laborious series of observations and experiments on the living habits of
insects. The community life of the social insects, in particular of the bees,
the development of the parasitic Hymenoptera, and the activities of leaf-
mining and gall-forming in ects may be specially mentioned among the sub-
jects dealt with by Reaumur in his great work — subjects to which he made
important contributions. Besides these his book contains a mass of valuable
detailed descriptions of larval and imaginal forms from practically all in-
sect groups.
Reaumur's physiological researches
Moreover, outside the sphere of insects he has presented biology with the
results of important discoveries. Thus, he has established the fact that the
shell of molluscs is formed by means of a secretive process, and in connexion
therewith he studied the formation of pearls in mussels. He studied also the
movements of a number of primitive animal forms, he investigated the elec-
tric phenomena in the ray, and he further observed the regeneration of the
extremities and other parts of the body of crayfish, in regard to this latter
phenomenon producing a theory reminiscent of Buffon's hypothesis as to the
body's being composed of organized particles. And, finally, he carried out
132. THE HISTORY OF BIOLOGY
several interesting experiments on the digestion, principally on the influence
of the gastric juices; he obtained gastric juice from a chicken by letting the
bird swallow a sponge attached to a piece of thread, with which the sponge
was after a time recovered from the stomach drenched with gastric juice,
which was afterwards used for the purpose of acting upon various kinds of
food substances. Among his contemporaries he justly enjoyed a great reputa-
tion; Linnasus cites him often and with recognition, while he had many
pupils, of whom de Geer in particular at once carried on his work.
Charles de Geer was born in 172.0 at Finspong, in the Swedish province
of Ostergotland. He was a descendant of the rich merchant and manufacturer
Louis de Geer, who had emigrated from Holland, and consistently with his
family's origin he received his education in that country. He studied at the
University of Utrecht, where he devoted himself to both physics and biology.
As a child he inherited Lovsta Foundry, in Uppland, and as soon as he came
of age he took over its management. He introduced a number of improve-
ments in the iron-manufacture and thereby acquired considerable wealth.
Regarded as one of the richest and most brilliant noblemen in Sweden, he
became in course of time Court Marshal and baron and received many other
distinctions. He showed great consideration for his workers, founding schools
for their benefit and improving their wage conditions. He was highly
reputed in scientific circles in Europe and was a member of several learned
societies. He died in 1778.
At an early age de Geer had been interested in entomology. In this field
he continued the investigations begun by Reaumur, and published under the
same title a sequel to the latter's great work, which it in every way equals
in value. It comprised seven volumes, containing observations upon the
systematic classification of insects, their habits of life and evolutionary his-
tory. Although contemporary with Linn^us, de Geer did not adopt his
nomenclature, but retained the old method of characterizing the species by
means of diagnoses. Otherwise he was a keen observer, who in more spheres
than one made contributions of lasting value, not least in regard to the
lower and hitherto neglected insect-forms.
Among the naturalists who during this period made valuable contribu-
tions to the knowledge of the lower animals should also be mentioned
Abraham Trembley (1700-84). He was born at Geneva, studied first of all
there, then in Holland and England, was for a time private tutor in certain
distinguished families, and finally became a librarian in his native town. His
reputation as a biologist is based upon his important monograph on the
fresh-water polypi. In this work he gives a careful account of a number of
"polypus-forms" — he includes both Hydra and Plumatella in the same
genus. He closely studied their habits, particularly their movements and
food, and was, properly speaking, the first who clearly realized their animal
SEVENTEENTH AND EIGHTEENTH CENTURIES 133
character. He observed their natural propagation, but above all he carried
out systematic and extensive experiments in regard to their power of re-
generation, thereby opening up for research a field that has been cultivated,
especially in modern times, with splendid results.
August Roesel von Rosenhof (1705-59) was born in Thuringia, but
worked mostly at Nuremberg, first of all as a painter an-d afterwards as a
naturalist. Under the striking title of Monafliche Insectenbelustigungen he
published in the seventeen-fifties a series of observations on the life of the
lower animals, illustrated with beautiful engravings done by himself. A
number of sound detailed observations regarding the life-habits and develop-
ment of insects are given in these writings, but he paid special attention to
the evolutionary history of frogs, from their mating and egg-laying through
all their larval stages, and has thus given to posterity valuable additions
to the knowledge of these creatures, which, as is well known, are much used
in modern experimental physiology. -
Pierre Lyonet (1707-89) was also a highly reputed biologist among his
contemporaries. Born at The Hague of French parents, he was given a very
extensive education; he was a brilliant linguist and at one time followed the
career of a diplomat. As a biologist he applied himself most actively to the
sphere of insect-anatomy, in the spirit of Swammerdam; an admirable work
of unsurpassed brilliance even in our own time is his great monograph on the
larva of Cossus lig?iiperda or goat-moth caterpillar, the anatomy of which he
studied and illustrated with extraordinary conscientiousness and keenness of
observation
CHAPTER X
EXPERIMENTAL AND SPECULATIVE BIOLOGY IN THE
EIGHTEENTH CENTURY
BESIDES THESE MONOGRAPHisTs, who Were highly regarded in their own
day and are still well worth reading even nowadays, there lived dur-
ing the eighteenth century many scientists whose works embraced
fields of research of wide extent in regard to both the material investigated
and the problems dealt with. In particular, experimental biology and theo-
retical questions in connexion therewith were developed on a considerable
scale during this period by scientists who have merited the attention both of
their own age and of posterity. Foremost among these should be mentioned
Haller, a great man in his own age and a scientist for all time, famous as
a botanist, anatomist, physiologist, statesman, and poet.
Albrecht von Haller was born at Berne in 1707. His father was a
wealthy and highly reputed lawyer, who gave his son a thorough education,
at first in his own home with a private tutor, then at the University of Tub-
ingen, and finally at Leyden under Boerhaave. Young Albrecht was an infant
prodigy; at the age of ten he had a thorough knowledge of Greek and
Hebrew, at fifteen he had written an epic poem and some tragedies, at nine-
teen he was a doctor of medicine. It was obvious that a young man thus
equipped would in time become something quite out of the ordinary; un-
fortunately, as so often happens, none of the successes that he actually at-
tained fully reached the height of his dreams. Having taken his degree,
Haller studied for a time in Paris, afterwards settling down in Berne as a
physician. He there became universally known as a botanist and poet and in
1736 was appointed a professor of medicine at the then newly-founded
University of Gottingen, where he did splendid work; he laid out a botanical
garden and built an anatomical theatre, founded a still existing and much
thought-of scientific society, and besides found time for scientific author-
ship of an extraordinarily many-sided character. But he never won content-
ment; he was troubled with melancholy and a longing for his native country,
and finally he resigned his professorship and returned home to Berne (in
1753). There he was elected to the municipal council and made a name as
a distinguished official; his services were utilized as a diplomat and he per-
formed his duties in that capacity with honour. Meanwhile he continued
his scientific writing with undiminished zeal; his productivity was nothing
^34
o
2
w
H
C/5
D
<
O
U
O
f«^ /-^
fpii
ay
.^^
SEVENTEENTH AND EIGHTEENTH CENTURIES 2.35
short of amazing; a list of his writings gives a total of 650, among them
many very extensive works. But during this period the melancholy that
had pursued him since his youth increased. He felt dissatisfied with the re-
sults he obtained, as also with the new ideas that were becoming more and
more common. In particular, the increasing free-thinking in the world
troubled him and called forth a number of pamphlets in defence of Chris-
tianity from his hand. He himself, however, in spite of his firm belief in
the Gospel, had no internal peace, but ruminated over the Tightness of what
he had done in his life; the vivisections which he had performed, and which
always troubled his sensitive mind, now appeared to be specially repugnant
to him. After some years of decline in health he died in 1777.
In his youth Haller devoted himself principally to botany and verse-
writing. This, of course, is not the place to criticize his poetry; this much,
however, may be mentioned, that he is considered to have discovered the
poetical value of Alpine beauty; otherwise his poems are now read apparently
only by students of literature. As a botanist Haller appears in conscious
rivalry with Linnasus; he seeks to set up, in opposition to Linnaeus 's arti-
ficial system, a natural system based primarily on the character of the fruit.
It was not successful; Linnxus's investigations into the possibilities of the
natural system clearly proved that the time was not yet ripe for such a one,
whereas the sexual system in every respect fulfilled the requirements of the
period. Haller was embittered by defeat; and although his criticism of his
successful rival may be partially justified, nevertheless his disappointment
over his failure is clearly apparent. 1 In actual fact the two rivals were in-
commensurable; in contrast to Haller's magnificent but divided many-sided-
ness is Linn^us's consummate limitation — the former never touched
supremacy at any point; the latter possessed only one sphere, but there he
was master.
Haller' s physiological researches
The branch of biology in which Haller made his finest contribution is un-
doubtedly physiology; in this field he has not only developed the method,
he also established new and important facts, made valuable additions of a
purely theoretical nature, and finally compiled the results hitherto attained
in a comprehensive manner, which should be a pattern for all time. His
writings on this subject consist partly of a mass of articles written for jour-
nals, in which he recorded the results of his direct observations, partly of an
immense physiological compendium, and, finally, of a smaller and extremely
concise handbook on physiology, which was still in use for educational
purposes until the nineteenth century. In the foreword to the last-mentioned
1 Curiously enough, Haller published his bitterest attacks against Linnxus in the form of
a series of disputations (Dubia exLinnai fundamentis hausta, Gottingen, 175 1-3) which purported
to have been written by his son, a young medical student.
Z36 THE HISTORY OF BIOLOGY
work he defines physiology as " animata anato??ie," a vitalized anatomy, and
it is in fact the phenomena of life that he seeks to discover in his special
investigations. The most remarkable of these is without doubt his treatise
on the irritable and sensible parts of the body, which he published simul-
taneously in several languages. In this investigation he first establishes the
fact that the organs of the body are partly irritable, partly non-irritable;
why this is so, science cannot discover; it can only show that it is so. As
irritable (jnitahilis) he mentions such a part of the body as contracts upon
being touched; as sensible (sensibiUs), again, he defines a part of the body,
contact with which induces an impression in the mind. Which organs belong
to the one or the other category is a question which can be answered
only by experiment. The performing of such experiments on live animals
Haller finds highly revolting, but in the interests of truth it cannot in this
case be avoided. Thus he has proved that, of the two layers of skin, the
epidermis is non-sensitive, the cutis on the other hand has feeling, and adi-
pose tissue is non-sensitive. Muscles are sensible, but this is due not to the
actual muscular substance, but to the nerves which are in connexion there-
with; the tendons, again, are non-sensible, because they are not connected
with nerves. Bones and periosteum are insensible, the cerebral membrane,
the peritoneum, and the veins likewise. The intestines are sensible, but not
the liver, the spleen, or the kidneys. Irritability exists in muscles, but is
induced through the nervous system; thus the diaphragm has been made
to contract by irritating a severed nerve. Therefore the irritability cannot
have anything to do with the mind, for that is indivisible. Haller then enu-
merates a number of irritable organs — veins, intestines, sexual organs. Fi-
nally he discusses the question of what it is that induces irritability. Muscle
is composed of lime and earth; if it is asked which component part is irritable,
the answer must be the lime. Lastly he deals with the question of vital
organs, which serve the unconscious manifestations of life, and voluntary
organs, which serve the will.
The investigation here referred to must without doubt be regarded as
one of those that have led biology into new directions. Not only the scheme
of the work, but also the method of presentation he employs and the con-
clusions he draws are each of fundamental importance; irritability and sen-
sibility are facts, which hold true to this very day,^ and the experimental
method by means of which the phenomena have been established is still
used today. There occur, indeed, in the examples just given one or two actual
mistakes — thus the peritoneum is, as a matter of fact, sensible and the in-
testines insensible — while the actual theoretical treatment suffers from the
^ Instead of irritability the characteristic property of muscle is nowadays termed "con-
tractility." Haller's irritability theory was later applied without distinction to various organs
in the body, thereby causing considerable confusion.
SEVENTEENTH AND EIGHTEENTH CENTURIES 137
fact that Haller did not succeed in producing any term for tissue — muscles,
intestines, and other viscera are quoted as organs of the same category.
Moreover, the chemical basis of his muscular theory, the composition of
lime and earth, is extremely primitive. But in spite of all this Haller, as
a result of his work in this field, has won a brilliant name in the history
of science.
His comfendiums of physiology
The compendiums which Haller produced still further extended the services
he has rendered to the development of biology. In the two works on general
physiology cited above he has summarized all the then known physiological
facts in a concise and easily accessible form. He starts by taking the simplest
component parts of the body, which are divided into solid and fluid. The
simplesi^elements of the solid components are, according to him, the fibres,
the composition of which has already been mentioned — lime and earth.
By cell-tissue, a word which often occurs, is meant what modern histology
terms adipose tissue. Haller considers the most vital part of the organism
to be the blood-vessel system; it represents the element that connects together
his whole physiological theory. In his description of each organ he always
starts with its blood-vessels. The more blood-vessels an organ has, the more
important it is. Of the thyreoidea he says that we do not know its fuxiction,
but it must be an important one, since the organ in question is so rich in
blood-vessels. Out of the blood are produced all the fluids of the body in
an entirely direct manner; thus he claims to have found direct passages be-
tween the arteries and the salivary ducts in the salivary gland; even the
lymph he believes to emanate from the arteries. The purpose of respiration
is to give the blood warmth. Haller was keenly interested in the structure
of the brain, but the results he attained are not to be compared with those
gained by Swedenborg at the same time on purely speculative lines. Haller
has only vague ideas on the cerebral cortex; the medulla is the most vital
part of the brain, and, in his view, the nerves are filled with a fluid which
gives rise to mental impressions. Towards many of the biological points of
dispute of his own time Haller tries to adopt a somewhat neutral attitude;
as, for instance, in the dispute between the ovists and the animalculists,
in which, however, he sided on the whole with the latter, since he held
that the spermium — " vermiculus seminalis," as he calls it — is the origin
of man, just as the larva is that of the fly. On the other hand, he describes
the follicle, or the egg, as he, like his contemporaries, calls it, as partaking
in the production of the embryo. As regards the question of preformation
or epigenesis, he is on the side of preformation. On the whole, he gives a
conscientious account of such views as he himself does not accept and dis-
plays in these works both creditable impartiality and a universal knowl-
edge of literature. He has taken special advantage of the latter quality in
138 THE HISTORY OF BIOLOGY
his bibliographical works — Bibliotheca anatomica, botanka, and chirurgica —
in which he has compiled information regarding all the literature published
till then in various spheres of science. These "bibliotheca" are even in modern
times of importance to the student of scientific literature and are remarkable
for their completeness, though also unfortunately for the mass of misprints
which mar their utility.
Mailer's refutation
Haller has been very diversely judged. On the one hand, he has been re-
garded both by his own age and by posterity as the foremost anatomist
and physiologist of his century and as the founder of modern experimental
physiology, while on the other, as has so often been the case with scientists
of many-sided interests, he has been accused by specialists of unreliability
in points of detail. His great service, particularly to the development of
physiology, can, however, never with justice be denied; his experimental
method and its results are undoubtedly of fundamental significance. As re-
gards his general conception of life, on the other hand, Haller has to a
certain extent stood at an old-time view-point, just as in his writings he
summarized the results hitherto attained. This to some extent explains how it
came about that the immediately succeeding age picked a quarrel with him;
thus, Goethe finds fault with him for his views on the limitations of the
knowledge of nature, which, it is true, are but little in accordance with
natural-philosophical speculation; but above all he fell foul of a contem-
porary scientist who, starting out from an entirely different standpoint and
having different preconceptions, arrived at an entirely opposed fundamental
view in regard to science — namely. La Mettrie.
JuLiEN Offroy de La Mettrie was born in 1709 at Saint-Malo in Brit-
tany. His father was a wealthy merchant, who had his son brought up to be
a priest. He studied theology in Paris and there joined the Jansenistic sect, a
movement in the French Church known for the strictness of its rules and
ideas, but disfavoured and persecuted by the Government. A physician in
his native town, however, succeeded in awaking in the young theologian
an interest in his profession, and so it came about that La Mettrie began
to study medicine, first in Paris and then at Leyden under Boerhaave. Having
passed his examination, he set up in practice for a time in his native town
and then became physician to a regiment of guards in Paris; by that time
he seemed to have prospects of making a brilliant career, as he was well
known both for his successful cures and as a witty and refined man, with
social aptitudes. But these high hopes soon had to be abandoned. He had
begun his scientific writing by translating into French some of his master
Boerhaave's more important works. This was viewed with disfavour by the
high-conservative medical faculty in Paris, which had consistently opposed
Boerhaave's theories, just as at one time it had opposed those of Vesalius
SEVENTEENTH AND EIGHTEENTH CENTURIES 139
and Harvey. And La Mettrie still further increased their irritation by pub-
lishing a number of satirical pamphlets against his opponents. Eventually
the latter found an opportunity of taking their revenge; in a work, L'His-
toire naturelle de I' dme, he had expressed views that were considered to be
at variance with the Christian faith. The theologians rushed to battle and
La Mettrie's friends advised him to go to Holland until the storm should
blow over. At Leyden, however, he printed a new pamphlet, which still
further aggravated his position; this was the famous treatise L' Homme ma-
chine, which was published anonymously, it is true, but which was immedi-
ately recognized. Its contents were such that the author could by no means
count upon even Dutch tolerance; he had to take precipitate flight and to
remain in hiding for a time. His fate would now have been deplorable had
not Europe possessed a reigning monarch who was absolutely indifferent
to religious problems, but who, on the other hand, was amused by witty
companions; this was Frederick II of Prussia. La Mettrie was summoned to
Berlin, was appointed lecturer at the royal court, and besides was given an
opportunity of practising as a physician. He enjoyed these privileges only
for a space of three years; in 175 1 he died as the result of an accident. He
had always boasted with some pride of his power of enjoying life's pleasures
both qualitatively and quantitatively; so at a feast, just to show off, he
ate enormous quantities of truffle pasty, immediately fell ill, and died in
terrible pain; probably the pasty had contained septic poison. This tragic
end, however, still further increased the ill fame caused by his writings;'
his name has, in fact, been one of the blackest in the whole of the eighteenth
century. In many respects, however, he paved the way for ideas which mod-
ern biological research has adopted and it is therefore worth while paying
some attention to his views.
La Mettrie's polemical ivorks
In his writings La Mettrie shows himself to be a marked oppositionist.
It is destructive work that amuses him most, and he likes best to pit his
strength against what his contemporaries regarded as the most unshakable
foundations both of the knowledge of existence and of the social order and
good manners. His polemical writings are sometimes brutally frank, some-
times subtly insidious, but he invariably challenges deep-rooted ideals, both
scientific and traditional, and is quite prepared to call white black and black
white. His love of truth goes just so far as serves his immediate purpose,
but undaunted courage we cannot deny him, and he has a firm faith in the
^ In ancient times it was held to be a matter of fact in High-Church quarters that no one
can die in peace without the Church's blessing. In this connexion it has been related that Luther
hanged himself (a Catholic statement), that Spinoza died under the influence of opium, and
Voltaire in a fit of madness. In furtherance of this kind of propaganda La Mettrie's above-
described death, which is historically confirmed, was a good find indeed.
X40 THE HISTORY OF BIOLOGY
subject that he made his own. What his writings seek to create is a general
cosmic view based entirely upon "philosophical" — that is to say, natural-
scientific — principles, for to him philosophy and natural science are iden-
tical. In contrast to the teachings of theology, politics, and the morality
that is based on them, he wished to create another ideal of justice and virtue
based on "natural" principles, and in contrast to the explanation of life
and nature which priests (and "philosophers" dependent upon them) have
given in support of ancient authorities, he would set up another explana-
tion, founded upon direct observation of the phenomena of life. He is thus
the first to enunciate a purely natural-scientific view of life, and in doing
so became the precursor of many similar endeavours in our own time. Herein
lies his greatest originality, for in most of his subjects he merely sets forth
in detail observations recorded by others, and his writings can hardly be
said to possess scientific form in the stricter sense of the term; they are pam-
phlets published for agitational purposes, often more likely to persuade than
to prove. Of these the two which have been cited above won the greatest
notoriety; among his other publications there is really only one — entitled
Systeme d' Epicure — that is of any great interest.^ His work on the natural
history of the soul, published before he had finally broken with his native
country, maintains a somewhat cautious tone and is therefore written in
a more scientific form. UHomme machine, again, is nothing but a piece of
agitation, and Systmie d' Epicure is a collection of aphoristic contributions
to a general knowledge of nature. The view of the functions of the human
body on which La Mettrie bases his speculations is by no means a new one;
it is the mechanistic theory of the bodily functions, which, founded by
Descartes, had been developed by Borelli and Perrault, by Hoffmann and
Boerhaave. To their observations La Mettrie has but little to add; the most
valuable contribution is an exposition of the independence of the vital func-
tions in the various parts of the body, confirmed by observations of the
manifestations of life in detached organs and bodily parts even in the highest
animal forms. His theory of fertilization may be worthy of mention; he
believes that one single " j-^^^rw^-animal " penetrates each ^gg and is there
further developed into a new individual; he thus belongs to the animal-
culist party. But it is not the life of the body that interests La Mettrie most;
it is the functions of the soul that forms the chief subject of his literary pro-
duction. With regard to the soul his views are quite clear; it does not exist,
or at any rate not in a form that the theologians and suchlike would have
it. As a matter of fact. La Mettrie is not quite certain what the soul is;
■* There would be no point in dealing herewith La Mettrie's strictly medical writings; of his
philosophical treatises, L'Hommt plantt is a development, driven to absurd lengths, of the com-
parison between the vegetable and animal organs which Linnasus had already arawn up; again, Lts
Animaux plus que machines contains little more than answers to his opponents' accusations.
SEVENTEENTH AND EIGHTEENTH CENTURIES X4I
sometimes he goes so far as expressly to point out that its essence must al-
ways remain unknown, while, on the other hand, he is quite convinced as
to what it is not: it is not the immortal spirit, distinct from the body, that
it is officially declared to be. During a fever La Mettrie had observed how
the faculties of the soul within him were affected by the course of his sick-
ness, and in his medical practice he had remarked the same phenomenon
in many of his patients. By thus making the influence of the bodily functions
upon the intellectual life a subject for investigation and even experiment
La Mettrie discovered that field of research which in modern times is termed
psycho-physics and which has been so successfully investigated by research-
workers with what are, in principle, the same methods as his, though with
an entirely different standard of scientific criticism. La Mettrie, in fact,
suffered the usual fate of a pioneer in not being able to free himself from
the prejudices he attacked.
His general conception of the human soul
He begins by accepting the old division into a vegetative, a sensitive, and
a rational soul; he analyses the first two and finds that their functions are
dependent upon those of the body, which indeed his predecessors had also
taken for granted. He devotes the main part of his investigation into the
soul to trying to discover the operation of the sensitive soul; he gives an
account of mental impressions and their mechanisms, in the course of which
he makes several striking observations, inter alia regarding the subjectivity
of mental perceptions. He discusses the localization of the mental functions
in the brain with extraordinary keen-sightedness and thence goes straight
over, with a somewhat daring mental jump, to ideas, which he treats — very
naively — as bodily entities, the grandeur of which he tries to estimate.
After having thus converted ideas in general into bodily phenomena, he dis-
cusses in connexion therewith a number of such ideas — memory, imagina-
tion, talent, etc. — all of which are to him likewise material, so that finally
there is nothing left of the rational and immortal soul that the theologians
have made it their mission to cherish. Thus he accumulates a mass of evi-
dence to show that the soul of man is fundamentally the same as that of the
animal; he cites examples of animal affection, gratitude, and such feelings,
and seeks, on the other hand, to adduce proofs that man possesses animal
qualities. He quotes in all seriousness a number of miraculous stories of
human beings who have lived like animals in the forests — probably em-
broidered tales of runaway lunatics — he describes the orang-utan with the
entirely human characteristics which were at that time ascribed to that ani-
mal, and hopes that it will be possible to teach it to talk by a method of
teaching the deaf and dumb to speak which had just been invented.^ And
^ Why the deaf-and-dumb method should have to be used for an ape which can hear just
as well as a man is not explained; probably it was the novelty of the method that made it seem
so wonderful and induced the hope of its performing further miracles.
X42. THE HISTORY OF BIOLOGY
finally he expounds a quite extraordinarily childish theory regarding the
"natural" origin of man and the living creatures here on earth.
The ' ' natural ' ' origin of man
Starting from Buffon's above-mentioned idea of the living particles scattered
about in space, out of which all living creatures have arisen, he assumes that
similar particles, intended to form human beings, have accumulated in the
earth and given rise to a number of human individuals, some defective, others
perfect. To the question why the earth no longer produces human beings
in this wise he would answer that the earth is now old and weary; in answer
to the question how the human babies thus produced eventually developed,
it may be supposed that they were brought up by kindly beasts of prey,
just as a small child had, it was said, recently been brought up by a she-bear
in Poland.
On reading such absurdities one recalls the days of old Empedocles, but
there is nothing to indicate that La Mettrie was not serious, as far as he
could be serious over anything. There is, it is true, no sign of scientific criti-
cism apparent in speculations such as these — in comparison with them
even Buffon's most daring assumptions are temperate and founded on facts —
but at least they have their interest as a sign of the times, and the endeavour
which finds expression therein points ahead to the "natural-creation sto-
ries" of our own day. No one before had ever dared so openly and so rashly
to break with the old, officially accepted, traditions, then upheld by the
whole authority of the State, and even among La Mettrie's contemporaries
there was no one who would have dared to abandon the belief in a God as
the Creator of the world and in the immortality of the soul; both were re-
garded as indispensable bases for even the most liberal-minded morality.
But La Mettrie would even reform morals and social life; he desired to create
a natural and philosophical system of ethics in place of the official theo-
logical and juridical system. Like others of his contemporaries, he believed
in man's natural inclination to virtue and happiness and he propounds
certain quite justifiable suggestions for reform. He would forbid wearisome
memorizing in the schools, maintaining that child education should be based
upon the exercise of the natural powers of observation, and he urges the
courts of law to differentiate between deeds committed by the mentally de-
ficient and ordinary crimes. The highest aim should be to make the world
happy, but the main point of the art of living that he preaches is first and
foremost "la volupte " — that is to say, in fact, sexual desire, the satisfaction
of which with the greatest possible enjoyment and the least possible risk
he discusses in a lengthy treatise, claiming thereby to lead humanity to the
height of rational worldly wisdom. This philosophy of licence became, as
is well known, widely popular during the latter half of the eighteenth cen-
tury; in itself it would of course have no concern with this present work
SEVENTEENTH AND EIGHTEENTH CENTURIES 143
had not several of its champions, like La Mettrie, based their philosophy
upon arguments derived from natural science, which contributed to bring
the latter into double discredit — both as hostile to revealed religion and
as promoting all kinds of flippancy and social disorganization. In this the
contempt shown for natural science in the age of romanticism, at least in
part, finds its explanation. And yet even La Mettrie possessed one feature
that is reminiscent of romanticism, or rather of the " Stunn und Drang'
period: he has left us a characteristic description of himself, in which he
talks of his kindly, innocent heart, which never committed any sin, even
though his thoughts did so, and he counsels us not to judge his morals by
his writings. There are, however, reasons for supposing that his life and
his teachings were not inconsistent with one another. He was certainly no
paragon of virtue, but he was a child of his age and he had ideas that were
in advance of those of that period.
On the whole, we find during the eighteenth century a great number of
philosophical speculations widely differing from one another; certain of
them broke boldly away from all the old traditional ideas, while others
sought to reconcile the old and the new. To the latter type belong in a
marked degree those attempts to explain the nature and development of
living organisms that were published by Bonnet, who in a remarkable way
combines ideas of value for the future with conceptions based on a cosmic
theory which had already been abandoned by most thinkers.
Charles Bonnet was born at Geneva of wealthy parents in 1710. The
family had emigrated from France at the time of the persecution of the Hu-
guenots. He studied law and was elected to the council of his native town,
but at the same time he evinced a lively interest in natural science and even-
tually devoted himself entirely to that pursuit. As a pupil of Reaumur he
applied himself chiefly to insect biology and in this field carried out work
of lasting value. A serious ophthalmic disease, however, soon compelled
him to give up making direct observations and all practical work of any
kind, so that, being a man of independent means, he spent the rest of his
days engaged in purely theoretical speculations in natural science and phi-
losophy. He died in 1793 on his estate in the neighbourhood of Geneva.
Even in his earliest works Bonnet shows himself, apart from his ac-
counts of actual observations, a natural philosopher, and in his last works
speculation alone predominates. As a thinker Bonnet is entirely in accord
with the Christian point of view, and his writings, by contrast to the free-
thinking that was so prevalent in his time, acquire a sharply polemical
and religiously fervent tone, with the result that sometimes even his purely
practical declarations are made in a form that sounds more like those of a
lay preacher than a scientist. His writings are extremely difficult for a modern
reader to appreciate; one has to search through scores or even hundreds of
X44 THE HISTORY OF BIOLOGY
pages full of enthusiastic praise of the Creator, of effusive outpourings re-
garding the life of the angels and the existence of the human soul in another
life, in order to find amongst it all biological observations of lasting value
and shrewd theoretical discussions on natural-scientific problems. Bonnet
bases the whole of this speculation, so widely at variance with the spirit of
his age, on Leibniz, who, we may remember, strove to reconcile the Christian
beliefs with the results of natural science and philosophical thought. Haller's
fervently Christian conception of nature also strongly influences Bonnet,
who, having sprung from the high-conservative and strictly Calvinistic
patrician class of Geneva, was perhaps even in his youth opposed to that
tendency to free-thinking which prevailed in most contemporary scientific
circles.
Bonnet's discovery of parthenogenesis
Bonnet takes his place in the history of biology primarily as the discoverer
of parthenogenetic reproduction. For years he studied the reproduction of
the aphides, and succeeded in establishing the existence of a number of sum-
mer hatches of females who without fertilization propagate by producing
live offspring; towards the autumn a new generation arises, this time con-
sisting of males and females, which mate, the females then laying eggs,
which hibernate. He also discovered and studied other peculiarities of insect
reproduction, as, for instance, the peculiar propagation of the pupiparous
flies. Further, Bonnet followed up with great care the study of the phenomena
of division and regeneration which Trembley had discovered; he observed
a large number of lower, colonizing animals, belonging to the Coelentera and
the Bryozoa, and experimented with them, as also with fresh-water Annel-
ida and common earthworms, observing not only the regeneration which
results in normal individuals, but also such as results in malformations —
in this respect a precursor in a special field of research, which has been
very highly developed in modern times. He studied with great exactness
the metamorphosis of insects, endeavouring to discover what changes the
parts and organs of the body undergo in the process of evolution from larva,
through the pupal stage, to the imago; and at least as far as the intestinal
canal is concerned he made, on the whole, correct observations. Again, he
studied the adipose tissue and the part it plays during the metamorphosis
period of insects as reserve nutriment for the prospective individual. Bonnet
also experimented with plants and was one of the first to study their tropisms
and growth-movements. The whole of this valuable collection of facts,
however, he accumulated to form the basis of his theoretical speculations
upon life on the earth — one might even say, in the universe — which were
to him the most essential function of science.
SEVENTEENTH AND EIGHTEENTH CENTURIES 145
His preformation theory
Of Bonnet's scientific theories the best known is his thoroughly worked-
out preformation theory. His "incapsulation" theory, according to which
every female individual contains within her the "germs" of all the creatures
that originate from her, the one generation within the other, and that thus
the first female of every species contained within her all the individuals of
that species that have ever been produced and that will be produced until
the end of time — this theory is really the very foundation on which all
his biological speculation was built. He found actual support for it in his
observations of the reproduction of the Aphididre; in the parthenogeneti ally
produced, new-born female of the plant-louse he saw the ready-formed rudi-
ments of a new generation, and even the metamorphosing insect shows the
imago ready formed beneath the pupal skin. In the plant, on the other hand,
the germ and the cotyledons are visible in the seed, and the bud encloses
the leaves that are to emanate from it. He therefore considered himself fully
justified in seeing in these facts a universal law governing the whole of ani-
mate nature; he is strongly opposed to all epi genetic theories and character-
izes as legends the observations purporting to show that in the embryo of
the chicken certain organs are developed before others. However, it is never
made quite clear whether these germs, which thus exist in infinite numbers
incapsulated within one another, are to be regarded purely corporeally or
whether they are some kind of ideal entities after the manner of Aristotle.
Bonnet, as a matter of fact, draws quite a number of his theories from Aris-
totle, starting from that of the ultimate cause, God or the supreme intelli-
gence, and of the harmony and finality of the universe. At all events, the
germs exist not only in the ovaries of the females, but also, in some animals
at any rate, scattered all over the body. There is, in fact, no other way by
which Bonnet can explain how the bits of a cut-up earthworm are regen-
erated into new individuals; for the earthworm must, like all animals, be
assumed to possess a soul, and the soul is always one and indivisible; if,
then, the earthworm is to be regenerated, germs possessing soul-rudiments
must lie scattered throughout the body. Indeed, even a separate extremity
that is regenerated, as in the crayfish, for instance, must possess a separate
germ which is intended to replace it when it is lost, and the same holds
good for individual muscles and fibres, which are capable of growing again
even in the highest animal forms. — The whole of this germ theory is clearly
reminiscent of Leibniz's monad theory and thus has its origin in common
with both BufFon's and La Mettrie's doctrines of living particles filling the
universe; but while the two latter sceptics utilized the hypothesis to estab-
lish a theory of primal creation (spontaneous generation), thereby abandon-
ing the personal creator, the fervently religious Bonnet strongly repudiates
all idea of spontaneous generation and gives a number of reasons against
-L^G THE HISTORY OF BIOLOGY
it, which are in part valid at the present day — an instance among many
of how the scientist's personal conceptions influence his purely scientific
theories.
His descent theory
The preformation theory, however, represents only one side of Bonnet's
curious speculative investigations. One idea that occupies his mind quite
as much is the thought of the progressive development going on in nature.
His firm conviction as to the wisdom of the Creator has made of him an
incorrigible optimist; he is absolutely convinced that nature is advancing
towards a high goal; he believes that there are heavenly bodies in which
this development, which he expects that the earth will eventually experience,
has already been attained — in which the stones possess organic structure,
the plants are sensible, the animals talk, and men are angels. And just as
he expects an advance beyond the present stage, so he believes that this is
the result of a process of evolution; the "germs" that are incapsulated
within one another in an individual are not alike and never have been; on
the contrary, he expressly maintains that if one were to see a horse, a hen,
a snake, under the form they had when they first came into existence, they
would be unrecognizable. These changes he accounts for by a series of stages
of development which the earth has undergone and each of which has been
cut short by some vast natural catastrophe, which destroyed all living things,
but always spared the germs out of which fresh life -forms arose. The last
of these catastrophes was the one that destroyed the earth before the six
days of the Creation referred to in the Books of Moses, the historical authen-
ticity of which Bonnet was naturally careful to maintain, but which he
interprets somewhat freely, according to the orthodox view. As the result
of a coming catastrophe he expects the perfecting of the world's existence,
as indicated above, while he assumes from the presence of fossils in the moun-
tains a series of previous epochs of existence with living creatures that did
not resemble those now existing and from which these latter have originated.
Just as Bonnet manifestly bases this geological theory on Buffon, so he
is the precursor of both Cuvier and Lamarck in the same way; Cuvier's
famous catastrophe-theory corresponds too closely with Bonnet's to justify
the assertion that the similarity was accidental, while, on the other hand,
Bonnet in his express statements regarding the change that takes place in
species has forestalled Lamarck's descent speculations, though, it is true,
both the biologists mentioned succeeded in elaborating their ideas into a
far more perfect whole. But in still another respect Bonnet foreshadows these
two great pioneers of biological science: he maintains — again an obvious
connexion with Buffon — that nature draws no sharply defined lines between
the species, but that all life-forms on the earth pass into one another. He
draws up what he calls ' ' une khelle des etres naturels " — a series proceeding
SEVENTEENTH AND EIGHTEENTH CENTURIES Z^J
from the simple elements through the mineral kingdom, vegetable kingdom,
and animal kingdom in a long line right up to man. The transitions in the
series are, from the modern point of view, ingenuously chosen: the flying
fish provides the transition between fishes and birds; the ostrich, the bat,
and the flying squirrel between birds and quadrupeds; the polypus and the
sensitive plant between animals and vegetables. But then he also declares
that the whole of this division is only approximate and that perhaps the
series is not as uniform as he has made it; that perhaps molluscs and insects,
lizards and frogs do not follow one another consecutively, but are in reality
collateral with one another. Just as the long evolutionary series of Bonnet
clearly foreshadows Lamarck's evolutionary theory, so the assumption of
parallel evolutionary groups represents a first hint of the type theory that
Cuvier founded and whereby he reformed the entire zoological system of
classification and rendered possible the approach of the modern descent-
theory. And as has already been pointed out, the points of agreement are
certainly not accidental; Bonnet enjoyed a great reputation amongst his con-
temporaries and the immediately succeeding age and was diligently studied.
Cuvier in particular has expressed his warm admiration for him and recom-
mended his writings for careful study; and other contemporary biologists
certainly knew his works and were to some extent influenced by them.
In the foregoing have been mentioned those theories of Bonnet that
have proved to be the most vital for the development of biology, and con-
siderations of space forbid a detailed account of all the shrewd utterances
which this imaginative man of genius scattered throughout his writings;
for example, his striking criticism of vitalism. In spite of his religious fa-
naticism he gives a purely mechanical explanation of the bodily functions
and cites the pointed objection to the vitalists — mostly Stahl and his
school — that "souls" are particularly convenient to have when it is a
question of explaining the phenomena of life; they do everything that is
asked of them and their non-existence can never be proved. Another time
he gives a detailed analysis of the different organs in the same body that
are dependent upon one another and shows how a change in one organ must
inevitably react upon the others; and on still another occasion he describes
his observations regarding different mental impressions — a problem which,
as is well known, Goethe made the subject of exhaustive study. Thus Bon-
net was a man full of ideas; and though in a great deal he must appear out
of accord with our age, yet undoubtedly many of his ideas are nowadays
incorporated in the general consciousness.
The experimental biological investigations that Bonnet made the basis
of his speculations were continued and considerably widened by Lazzaro
Spallanzani (172.9-99). Born at Reggio, the son of a lawyer, he studied law
at Bologna and at the same time took orders. He afterwards devoted himself
X^S THE HISTORY OF BIOLOGY
to natural-scientific studies and became professor of philosophy, first at Mo-
dena and later at Pavia. He applied himself to experimental research, particu-
larly in regard to regeneration and fertilization, and left all his predecessors
far behind him, in both method and results. In the amphibians, especially
salamanders and tritons, he found suitable subjects for the study of regenera-
tion even among the vertebrates, and he made as exhaustive a use of them
as was possible under the conditions in which he worked; he studied the
re-formation of the tail, extremities, and jaws, and this not merely for the
purpose of establishing the fact, but by means of dissection and microscopic
investigation he followed the re-forming of the various components of the
body: muscles, nerves, and bones. He observed the time that the regenera-
tion lasted and endeavoured to influence the process by means of altering
the conditions of food and temperature. He even experimented with the
phenomena of fertilization; by filtering the sperm of particular animals he
proved that the presence of the spermatozoa was essential if fertilization
was to take place; nevertheless he could not be induced to assume a direct
influence of these components upon the egg, but believed that the accompany-
ing fluid was the substance that stimulated the egg's development. For he
adhered as stubbornly as Bonnet to the preformation theory: he closely
studied the development of the frog's egg and followed the formation of
the backbone channel, but merely for the purpose of seeking evidence of
the entire animal's having been ready-formed in the egg. Eventually he be-
lieved that he had discovered incontestable proof thereof, when he saw the
frog's egg increasing in size within the body of the mother animal and before
it had been fertilized, and as growth is not possible without organs, the
larva of the frog must have been ready-made in the egg before fertilization.
Just as Spallanzani was thus convinced that he had found an undeniable
argument in favour of the preformation theory, another scientist published
a treatise which was to form the basis of a new conception of embryonic
development.
Caspar Friedrich Wolff was the name of the naturalist who led the
science of embryology into fresh paths. He was born in Berlin in 1733, ^^^
son of a tailor. He went through a course of medical training at the College
of Medicine there and thence proceeded to Halle, where he studied philos-
ophy after the system of Leibniz and his pupil Christian WolfF,^ and finally,
^ Christian Wolff (1679-1754) was professor of mathematics at Halle, whence he was
ejected through the intrigues of the Pietists (Stahl seems to have taken part in the persecution
directed against him) and then became professor at Marburg, but later he returned to Halle
and there worked as professor of philosophy. Under the general title Vernunftige Betrachtungen he
published a series of essays covering many different fields of human knowledge, in which he
expounds Leibniz's theories in a popularized form, in particular maintaining that everything
that happens must possess adequate reason for doing so, because otherwise something might
SEVENTEENTH AND EIGHTEENTH CENTURIES X49
for his doctor's degree, he published in 1759 the essay which was to make
his name famous. Having served for a time as an army surgeon, he applied
for and obtained permission to hold lectures in medicine in Berlin, which,
however, resulted in his coming into serious conflict with the professors
at the Collegium Medicum. Being of a peaceful disposition he was much
troubled at this and was delighted when he received a summons to St. Peters-
burg, where he became an academician and spent the rest of his life carrying
on his research work in peace. He died in 1794.
Woljf's generation theory
Caspar Friedrich Wolff is one of those who did not win fame until after
death. His own age paid little attention to him. Haller, to whom he dedi-
cated his afterwards famous treatise Theoria generationis, accepted the honour
in a friendly spirit, but paid little attention to the work, as also did other
biologists of the period. That Wolff was thus misunderstood by his con-
temporaries was due mostly to the fact that from the very outset he adopted
a course directly opposed to the then prevailing conception of the phenomena
of life; he began with a ready-made theoretical program, and the facts he
presents are collected for the express purpose of proving his already firmly
established convictions. By way of introduction he lays down the plan of
his work; by the body's '' generatio," or, as we should now call it, "evolu-
tion," is meant its creation Q' fonnatio"^ in all its parts, and its principle
is the force that brings about this creation. The upholders of the doctrine
of "predelineation" thus deny, he adds, that any "generation" takes place
at all. He starts, therefore, by declaring war on the preformation theory;
he does not base his rejection of it on the evidence of the facts he has ob-
served, but on purely theoretical reasons adduced by Christian Wolff's philo-
sophical methods. "He gives a true explanation of generation who derives
the parts of the body and their composition from the fixed principles and
laws governing them; . . . and he has perfected a theory of generation
who has succeeded in tracing the entire ready-formed body from these prin-
ciples and laws." The principles on which the fresh formation of organism
takes place are food and growth; food re-creates the simple components of
the organism, while through growth are formed entire parts of the body or
fresh bodies. Reproduction is, in fact, brought about by a "weakened growth
(yegetatio languescens)," whereby the newly-formed seed or embryo is sepa-
rated from the mother plant or animal and is prevented from growing further
in union with the latter. And that which produces all nourishment and
growth is, according to Wolff, the "inner force (vis essentialis^," a term
which he constantly uses to mean the ultimate cause of all that takes place
arise out of nothing, which is impossible. For the rest, he was a clever mathematician and, for
his age, a sound botanist, and contributed much towards inculcating an interest in natural
science in Germany.
X50 THE HISTORY OF BIOLOGY
in the organism, the idea of which he himself states that he borrowed from
Stahl. For just as the basis of Wolff's research work is the rejection of pre-
formation, so its final object is the abolition of "mechanical medicine" —
the theory that holds that the living body should be regarded and treated
as a machine, which finds the explanation of the phenomena of life in the
form and composition of the bodily parts, or, as it is expressly stated, in
anatomical principles. This theory Wolff declares to be a product of the
imagination and produces a number of arguments to prove its falseness.
Wolff, howe'' .:, never arrives at any properly worked-out vitalistic theory;
after all, he deals with the phenomena of the body along mechanical lines
and his "vis essetifialis" he does not identify, as Stahl did, with the soul. On
the whole, Wolff's theory is vague and inconsistent if we compare it with
Stahl's mode of thought, which is certainly hard to apprehend, but is never-
theless in its way loftily conceived. The most serious result of Wolff's phil-
osophical method, however, is that he fancies it capable of explaining
practically anything; with a couple of phrases he throws a bridge across
even the deepest abysses of natural science; he has a theory ready to hand to
explain even such phenomena as those in the face of which modern biology
has to be content with merely establishing the fact. In all this he is a pre-
cursor of the natural philosophy of romanticism, and it was, in fact, this
that eventually procured for his views the honour they deserved.
His cellular theory
Wolff's treatise deals with the development of both plants and animals in
a constant endeavour to find factors common to both. In his opinion, the
growth of plants is due to the inner life-force drawing up moisture out of
the earth through the roots and into all the various parts of the plants; at the
points of growth this moisture is collected in especially large quantities;
through evaporation it acquires greater density and forms cuticles, which,
through fresh supplies of moisture, assume the form of ampulla;, the walls
of which are further thickened by evaporation, and the new ampulla force
themselves in between the earlier ones, whereby the substance of the plant
is renewed. The plant's vesicular system is formed through the circulating
sap's hollowing out ducts in the vegetable substance, the walls of these
ducts being likewise thickened by evaporation. The plant is then formed
by these ampulla; and ducts through a system of growth-forms, the abstract
and involved details of which it would take too long to follow. As mentioned
above, the florescence and germination are caused by a weakened growth —
"The adequate reason why within a certain period frondescence ceases and
germination begins is a diminution of the supply of alimental sap at the
point of growth, as is at once seen from the very definition of growth."
This is Wolff's scientific adduction of evidence. Similarly it is proved that
the germination and embryonic development consist in a renewed growth
SEVENTEENTH AND EIGHTEENTH CENTURIES 151
induced by the " perfecfum nutrimentum" with which the pollen' provides the
seeds; lengthy arguments are brought forward to show why the pollen is
and must be the most perfect form of nourishment that exists. Sexual re-
production, then, is nothing more than a renewed growth.
The fundamental principles on which growth proceeds in the vegetable
kingdom Wolff discovers in detail in the animal kingdom; in the embryo
of the chicken, which is his only subject of investigation in animal embry-
ology, he finds reproduced the same phenomenon of growth; the inner
force derives nourishment from the yolk of the embryonic lamina, this ali-
mental fluid coagulating, as in the plant, into ampullar and ducts, the latter
here represented by heart and vesicular system. Here, too, the details of the
embryonic development, which is described much less fully than that of
the plant, are of no particular interest; here again Wolff gives full rein to
his speculative imagination at the expense of detailed observation. Such an
assertion as that no one has discovered anything with a powerful magni-
fying-glass that could not equally well have been observed with a lower
magnification is sufficient evidence of how his speculations are out of accord
with reality. And still worse is his habit of comparing the structure of plants
and animals in detail; his comparison of a plant's vessels with the arteries,
of its suckers with the veins, rivals in absurdity most of what had hitherto
been perpetrated in that sphere — which was by no means little.
And yet it is just through his comparison of plant and animal develop-
ment that Wolff made his most important contribution to biological history.
He was the first to compare the elements of which the plant and the animal
are composed, and though the details of this comparison are for the most
part incorrect, it was at any rate he before anyone else who pointed out the
ampullar -like structure — in other words, the cell-tissue — that is common
to both. He thus carried science a considerable step further along the road
marked out by Malpighi and his immediate successors.
His epigenesis doctrine
Wolff's second service to science is generally said to be his introduction
of the doctrine of epigenesis into biology in place of the preformation theory.
We have previously found that the epigenesis theory is actually older than
the preformation theory; even Aristotle was an epigenetic, and his doctrine
was promulgated without contradiction even by Harvey, whereas the first
champion of the preformation theory was Swammerdam. It was thus an
ancient theory that Wolff adopted, and from the very outset he made it his
own on purely theoretical ground; it was only natural, therefore, that his
observations should eventually accord with the preconceived ideas. But the
progress of science was facilitated by the fact that — whether with pre-
conceptions or not — he saw more correctly in his microscope than his con-
temporary preformationists; for their part, they considered an embryological
Z52. THE HISTORY OF BIOLOGY
Study to be in the main superfluous, since everything was ready-formed
before, whereas Wolff showed that in this sphere there was still an immense
amount to be discovered and investigated, thereby opening up fresh fields
of research, in which very successful work was done during the succeeding
epoch. It has frequently been said, however, that in this question Wolff was
entirely in the right and his opponents in the wrong. This view is utterly
at variance with historical facts. When it first arose, the preformation theory
not only was fully justified and compatible with the scientific standpoint
of the time, but, as has been pointed out above, also constituted a real
advance, whereas the epigenesis theory, as Wolff formulated it, certainly
shot far beyond the mark. He who denied to the undeveloped egg all trace
of organic structure would undoubtedly have found modern ontogenetical
research, with its strong emphasis on the orientation of the egg and its
various parts and with the maintenance of the immutability of the factors
of heredity, highly preformational.
His romantic conception of nature
It is, however, by no means in his epigenesis theory alone that Wolff shows
himself at variance with his age and foreshadows a new era; the whole of
his scientific matter and his entire conception of nature differ in a marked
degree from those of his contemporaries. He is, as has already been pointed
out, a precursor in the course which natural science took at the end of the
eighteenth century and which is termed natural philosophy; this course was
directed, particularly in Germany, towards an entirely new knowledge of
nature, with the utter rejection of both the aims and the means with which
natural research had been carried out up to that time. Biological natural-
philosophy, however, is only a link in a universal cultural movement of far
wider influence, which will be explained later on. But first we must devote
some words to the application of the experimental method to botany during
the eighteenth century, as well as to one or two anatomical and morpho-
logical scientists who were at work during the same period.
In the eighteenth century the science of botany was governed, far more
than animal biology, by Linnasanism. During the same period, however,
one or two naturalists employed in their study of the vegetable kingdom
methods other than the purely systematic; as a rule they worked in obscu-
rity and their results were appreciated only by succeeding ages. One excep-
tion to this was the English experimental scientist, Stephen Hales, whose
work was highly appreciated by his contemporaries; his writings were
translated into French by Buffon and into German by Christian Wolff. And
he was well worthy of the attention paid to him, for he is without doubt
one of the most remarkable biologists of the eighteenth century.
Hales was born in 1679 ^^ Beckesbury in the south of England. He was
of good family, and after studying theology at Cambridge he took holy
SEVENTEENTH AND EIGHTEENTH CENTURIES 153
orders in the Church of England. He held various posts and finally became
vicar of Teddington, a parish in Middlesex, where he died in 1761. He was
known as a zealous priest, who worked for the advancement of his parish
from both a moral and a material point of view, and besides found time to
devote himself to important philanthropical works, such as the improve-
ment of prison conditions, the administration of charitable societies, and
inventions likely to prove of benefit to mankind. By those who knew him
personally he was extolled for his kindness and simplicity.
At Cambridge Hales had been attracted to the study of natural science,
particularly physics, chemistry, and botany; indeed, during his undergrad-
uate days Cambridge was primarily regarded as Newton's town. He main-
tained this interest throughout his life; it thus occurred to him to try by
way of physics to discover the conditions of the life and growth of plants,
an idea that he realized after experimental studies lasting many years. He
published his results in the year 1717 under the title of Vegetable Staticks.
In his ability to organize biological experiments and to draw conclusions
therefrom he was excelled by none of his contemporary scientists and by
but few of those that have come after him; it has been possible even in mod-
ern times to apply his experimental methods with profitable results. In re-
gard to his general cosmic conceptions Hales was, naturally, in conformity
with his profession and his age, a pious Christian, but like his master Newton
he strove conscientiously to discover the law-bound mechanical processes
undergone by the phenomena be investigated; he would never involve him-
self in hypothetical explanations of the manifestations of life.
Hales' s quantitative experiments
What Hales wished to discover by means of his experiments was, first of
all, the renewal of substance in plants, both quantitatively and qualitatively.
It is above all his quantitative investigations that merited and also won
general admiration; he was the first to apply systematically and on a large
scale the exact method of physics to animate nature. By watering previously
weighed potted plants for a given length of time with a fixed quantity of
water, and by weighing the plant daily during that period, he found out
its water-consumption; then he measured the leaf- and stem-surface of the
plant and calculated therefrom the relation between the surface of the plant
and the quantity of w^ater that it absorbed daily. Similarly, by means of
measuring and weighing he calculated the quantity of moisture that differ-
ent plants absorb out of the earth through their roots, as well as the speed
with which the sap circulates in the interior of the plant; and finally he
proved that plants absorb air through their leaves and stems and not only
through their roots, as earlier botanists declared. He was quite specially
interested in the problem of the relation of the air to living creatures. He
definitely maintains that the air contains component parts which are ab-
^54 THE HISTORY OF BIOLOGY
sorbed by the plant through the leaves and are converted into solid sub-
stances. Likewise he states, though with less certainty, that light penetrates
the leaves and co-operates in the alimental processes in them. In these asser-
tions attempts have been made to find definite proof of Hales's genius, and
there is undeniably in them a brilliant guess at facts that were established
at a later period, but in regard to points of detail his speculations on the
properties of air are undoubtedly far more deficient than his quantitative
experiments. It is true that gas-chemistry had been but little developed in
his time, but it would seem that he scarcely took advantage of what actually
was known; he certainly cites Boyle quite frequently, but he evidently knew
nothing of van Helmont's gas-experiments. To him all gases are "air," both
that which arises from the dry distillation of wood and that which is formed
by treating lime with acid. In such circumstances it was inevitable that the
great trouble he went to in investigating the influence of the air upon vege-
table, and even animal, life was to a great extent in vain. What he achieved
as an experimenter, however, is quite enough to ensure for him considerable
fame in the history of biology, and it was to be long before science advanced
beyond his point of view. For this to happen there was required above all
else a reformation of the science of chemistry — which indeed actually took
place at the end of the eighteenth century. This will be described in a fol-
lowing chapter. During the latter part of his life Hales also worked at ex-
periments on animals, particularly in regard to the blood-circulation, and he
displayed in this sphere the same power of arranging experiments and draw-
ing conclusions therefrom that he showed in his botanical investigations.
He measured the blood-pressure in live mammals by introducing into a vein
a tube in which the blood was made to rise; he calculated the speed of the
blood-stream in the veins and capillaries from the volume of the vessels,
the rate of movement of the blood-mass, and the resistance of the walls.
To these investigations he added a quantity of notes on medicinal and hy-
gienic subjects, with particular reference to the injuriousness of alcoholic
liquors, for he was a keen supporter of temperance. This fact gives his
Hamasfaficks, as he called his investigations into the blood, a far more
motley character than his treatise on vegetable physiology; nevertheless,
even these investigations are of some value and he holds a place of honour in
the history of physiology.
Among the plant-physiologists who, after Hales, distinguished them-
selves during the eighteenth century there are one or two who carried out
important experimental observations regarding plant-reproduction, and who
deserve special mention.
Joseph Gottlieb Koelreuter was born at Sulz, Wiirttemberg, in 1733.
We know, on the whole, very little about his life; he seems to have studied
in Berlin and Leipzig and spent some time in St. Petersburg; in 1764 he was
SEVENTEENTH AND EIGHTEENTH CENTURIES 155
made professor in natural history and curator of the botanical gardens at
Karlsruhe. He died in 1806. Before his appointment to the professorship he
had already published the first series of notes in which he recorded the
results of his experiment with the artificial fertilization of plants. As we
have seen, Camerarius was the first to experiment in this field. Linnxus fol-
lowed in his footsteps, carrying out, it will be remembered, the hybridization
of plants, but not being otherwise an experimental naturalist-worker in the
true sense.
Koelreuter s experiments in plant-life fertilization
KoELREUTER was the first who exclusively applied himself to experimenting
with the cultivation of plants with a view to explaining their fertilization
and development. To begin with he investigated the act of fertilization
itself; he examined the pollen under the microscope and came to the con-
clusion that its fertilizing property is due to an oily fluid that it secretes;
on the stigma of the pistil he found a similar fluid and concludes therefrom
that fertilization consists in a union of these fluids, just as an acid and a base
form a salt. Of greater value than these speculations are his careful observa-
tions of the method of transmitting the pollen; he is the first to explain
clearly that certain flowers are invariably fertilized by insects, and he also
pointed out the part played by the wind in the fertilization of other forms.
Of greatest interest, however, are his investigations in connexion with hy-
brid formations, a problem to which he eventually devoted all his attention.
In this sphere he paved the way for a field of research that, as is well known,
has at the present day attracted the interest of both the scientific world
and the public more than most others. To start with, for a number of years
he crossed different types of tobacco-plants with one another, afterwards,
however, proceeding to other plant genera: pinks, aquilegia, verbascum, and
others. Moreover, he was able to vary his experiments and to observe the
results thereof; he carefully compared the hybrids with the parent individ-
uals and noted similarities and dissimilarities between them; he mated the
hybrids with their parent species and observed the reversion to similarity
with the latter; he fertilized the hybrids with one another and obtained
results that foreshadowed Mendel's famous observations; he likewise even
noted cases which would be regarded at the present day as mutations. How-
ever, he naturally did not succeed in utilizing theoretically the results of
his experiments; besides, his ideas of the actual essence of fertilization were
all too vague — he believed, for instance, that by fertilizing a species with
a mixture of its own and foreign pollen it would be possible to obtain a kind
of semi-hybrids, which would be somewhat, but not very much unlike the
mother species. Further, he mixed his speculations up with certain mystical
ideas, particularly in the sphere of alchemy; he expressly compares the change
that the characters of the species undergo in hybridization with the con-
156 THE HISTORY OF BIOLOGY
version of metals effected by the alchemists, and presumes that, just as a
vegetable species can, by repeated unilateral crossing, be transformed into
another, one day man will learn to convert, by a necessarily gradual process,
one metal into another; in further evidence of which he finds a correspond-
ence between the pollen and the sulphur of the alchemists, proved by the
fact that pollen can be used as a means of reducing metal oxides, though this
is easily explained when the pollen is burned to coal, which has a reducing
effect. Moreover, the female sexual product is, in his opinion, "mercurial."
It is thus in the sphere of practical experiment that Koelreuter's greatness
lies; in this he is a pioneer, and his experiments in crossing were justly taken
as a model until Mendel's far more deeply thought-out experiments became
known. Koelreuter shared the fate of the latter, the greatest of his succes-
sors, in his works' being for a long time entirely neglected; it was not until
long after his death that they were rescued from oblivion and accorded the
appreciation they deserved.
The same fate of being neglected by contemporary and immediately suc-
ceeding ages was suffered by Christian Conrad Sprengel, whose investi-
gations into the fertilization of flowers were carried out in association with
Koelreuter's. Born in 1750 at Brandenburg, the son of a clergyman, Sprengel
studied theology and languages, afterwards devoting himself to the tutor's
profession. He was for some years a schoolmaster in Berlin and later became
rector at Spandau. After several quarrels with his superiors, his pupils, and
their parents he was dismissed with a pension in 1794 and then lived in
Berlin in a solitude that increased year by year until he died in 1816. His
irascible temperament contributed both to his failure as a teacher and to
his subsequent isolation; it was aggravated by the utter lack of understanding
with which his contemporaries received the results of the botanical re-
searches that had represented the chief interest of his life. He was unable
to print his last botanical writings, with the result that during his last few
years he applied himself to philology, apparently with but little success.
Sprengel' s experiments on plant-jertilixation
The work which at last brought his name to the knowledge of posterity
is his Das entdeckte Geheimnis der Natur im Ban und in der Bejruchtting der Blumen,
published in 1793. Under this somewhat pretentious title he collected a large
number of observations in connexion with the florescence of plants, and on
them bases a general theory of fertilization in the vegetable kingdom, which
in its essentials still holds good today. In conformity with his theological
upbringing he was fully convinced of nature's having been preconceived by
the wisdom of the Creator down to the minutest detail, and he consequently
set about trying to discover for what useful purpose the different parts and
properties of the flower were intended. As a result of his inquiries into this
subject he found, to begin with, that the flowers' nectaries are always pro-
SEVENTEENTH AND EIGHTEENTH CENTURIES 157
tected from the rain, and further that they are often characterized by special
colours, whence he concludes that their object must be to attract insects
to the flowers; but then the insects must themselves have some object in
their visits, and this he found, as did Koelreuter before him, to be the
conveyance of pollen from stamen to pistil. He now studied in detail the
relation of the insects to the flowers and noticed that certain flowers are in-
variably fertilized by special insect forms, others again by several different
forms, and that the position of the nectaries in each flower is adapted not
only to the flower's general conditions of life, but also to the insects that
visit it. Further, he discovered that in a number of bisexual flowers stamen
and pistil actually develop during different periods, and that therefore the
flower cannot be fertilized by its own pollen, but that pollen is conveyed
by the insects from flower to flower. This fact he calls dichogamy, a name
which is still used, and he concludes from it that "Nature does not appear
to desire that a flower be fertilized by its own pollen." Finally, he explains
more lucidly than any of his predecessors the contrast between flowers fer-
tilized by insects and those fertilized by the wind; on this subject, too, he
makes many striking observations.
Space forbids a more detailed account of the numerous shrewd and far-
reaching observations which Sprengel adduces in support of his theories.
Through his work he has laid a lasting foundation for one of the most
important sections of vegetable biology, and besides, in regard to insect
research, he has pointed out a method of far greater theoretical importance
than mere classification and collecting. And so the utter lack of understand-
ing shown for his work by his own age was all the more tragic. The natural
philosophers of the Romantic Age deeply despised detailed research work of
this kind, and the succeeding generation, which endeavoured to revive the
mechanistic conception of nature of the eighteenth century, felt embarrassed
by the detailed finality which Sprengel sought and found in the structure
and life of the flowers. It was only Darwin's authority that succeeded in
rescuing Sprengel from oblivion; in the flowers' and insects' mutual depend-
ence upon one another he found support for his theory of selection and
himself carried out investigations in this field, which will be described in
a later chapter. Thus Sprengel found redress — tardy but glorious.
CHAPTER XI
DESCRIPTIVE AND COMPARATIVE ANATOMY IN THE
EIGHTEENTH CENTURY
THE ANATOMICAL SCIENCE of the eighteenth century appears to be a
direct continuation of that of the previous century; no important
discoveries of a pioneer character were made, but those fields of
research that had been won were well investigated in detail, and this field
of inquiry can show names that testify to praiseworthy endeavour, if not
so much to brilliant genius. Of these names some of the more representative
will be mentioned in the present chapter.
Bernhard Siegfried Albinus was the son of a German physician of
repute who, after having studied at Leyden, held various posts in his native
country, but eventually returned to Leyden as a professor. Young Bernhard,
who was born at Frankfurt an der Oder in 1697, was consequently brought
up at Leyden and spent his life there. At the early age of twenty-four he was
made professor of anatomy and surgery, and lectured on these subjects and
on physiology until his death, in 1770. He was highly esteemed by his con-
temporaries and honours of many kinds were bestowed upon him. He was,
in fact, a thoroughly educated scientist and was gifted in many ways. He
was interested in the history of science and published critical editions of
the works of the leading anatomists — those of Vesalius, Eustacchi, and
Harvey were reprinted by him. His own works were extensive and profound.
He studied with great care the bone-structure of the human embryo and its
development, and even in the full-grown man it was mostly the bone-
structure and the musculature that interested him. He compiled a fine set
of engravings illustrating these two organic systems — Tabula sceleti et mus-
culorum corporis humani — a gigantic work in contents and weight, in which
in a series of splendid copperplates, drawn and engraved under his instruc-
tions by the famous artist Vandelaar, he reproduces the human bone-structure
and musculature in every detail. This work, which cost him a whole fortune,
is of its kind still unsurpassed. Besides doing research work Albinus also
practised as a doctor, and, thanks to him, Leyden still continued during
the eighteenth century to be a centre for anatomical studies.
One of Albinus's most brilliant pupils was Johann Nathanael Lie-
berkuhn. Born in Berlin in 1711, he was destined by his father for the
priesthood, and for several years had to study theology against his will.
^58
SEVENTEENTH AND EIGHTEENTH CENTURIES 159
After his father's death he took up medicine, studying first in Germany and
then under Albinus at Leyden, where he took his degree with a treatise
entitled De valvula colt. After paying visits to England and France, he settled
down in Berlin as a practitioner. He died in 1756. His short life and his
extensive practice prevented him from benefiting science as much as he other-
wise would certainly have been able to do, yet what he did achieve ensures
to him a place in the history of biology. He was above all an excellent
technician. He himself made microscopes of splendid workmanship and was
able to prepare under the microscope the most minute organic details.
Equally remarkable were his injections; preparations made by his own hand
are still preserved in the anatomical museum in Berlin. In fact, he used the
microscope for the purpose of studying injection-preparations, a thing which
had never been done before. His only really important work is his exposition
of the structure of the small intestine, in which he describes the Lieberkiihn-
ian crypts (called after him), as also those cells existing at the bottom
thereof, now called the Panethian cells, whose glandular nature, however,
he failed to discover. The whole work bears witness to his technical skill
both in injections and in microscopy, and forms a valuable contribution to
the development of microscopical anatomy.
Another pupil of Albinus's, who won a far greater reputation in his own
age, was Petrus Camper. He was born at Leyden in 17x2., studied there, and
took degrees in both philosophy and medicine. Having spent a couple of
years travelling, he was appointed professor at the academy at Franeker,
a small provincial university which, when he first went there, had only
four medical students, a number which he succeeded in increasing many
times over in a very short time. After five years, however, he obtained a
professorship in Amsterdam, and some time later one in Groningen, but he
finally gave up teaching and settled at The Hague, where he became a member
of the state council and took part in its political life. He died in 1789.
Camper is described as a man of an extremely superior personality,
brilliantly gifted, but quick-tempered and despotic. In his own time he was
regarded as one of the leading scholars in Europe and attained a splendid
position, both socially and financially. His many-sidedness was extraordi-
nary, almost reminiscent of Olof Rudbeck's. Besides carrying out anatomical
research in a number of different fields, he was a surgeon and gynaecologist,
hygienist, and expert in medical law and veterinary surgery, and in all these
spheres he made valuable contributions. He was, besides, an excellent draughts-
man and a leading connoisseur of art. He took measurements of the facial
angle in human beings of different ages and different races, and in comparison
therewith in higher vertebrates, with results of interest both to the history
of art and to natural science. This facial angle, which still bears Camper's
name, is formed by two lines, the one extending through the opening of
z6o THE HISTORY OF BIOLOGY
the ear and the bottom of the nostril, the other at a tangent to the most
protuberant part of the forehead and the chin. When Camper expounded
this idea before the Amsterdam Academy of Painting, with a view to giving
the artists a more accurate conception of the human form, he little thought
that in doing so he was laying the foundations of an entirely new branch
of science — modern craniology. In close connexion with this interest in
the structure of the human body are his special investigations of the apes,
particularly of those resembling man. He had procured as many specimens
as he could possibly get of the orang-utan, at that time extremely rare in
Europe, and he not only dissected a number of them, but closely studied a
live specimen. As a result of especially careful investigations into the mus-
culature of the extremities and the structure of the larynx, he proved con-
clusively that the animal is unable to walk upright, as La Mettrie and other
"philosophers" at that time imagined; nor can it in any form pronounce
an articulate language. The philosophers, however, were certainly far too
firm in their belief to allow themselves to be convinced by anatomical proofs,
all the more so as it could be urged against Camper that he was in all respects,
both religious and political, a conservative man.
Camper s anthropological and comparative-anatomical investigations
Camper was particularly interested in the anatomical investigation of un-
common and rare animals. He published monographs on the elephant, the
rhinoceros, and the reindeer, anatomically useful specimens of which he
succeeded in procuring owing to Holland's extensive shipping-communi-
cations. Of more general interest than these special researches is his study
of the bone-structure of birds, in which he describes for the first time how
the bones are filled with air to facilitate flight, and, in connexion therewith,
the air-sacs in the body which serve the same purpose. Of immense general
interest also are his comparative investigations into the auditory organs of
fish, whales, and reptiles, wherein he discusses the reproduction of sound
in various media and the ear's adaptability thereto, at the same time making
a close study of the different parts of the auditory apparatus. Finally, Camper
carried out an anatomical investigation of a highly original kind in his
essay "On the Best Form of Shoe," in which, after a detailed description
of the bone-structure of the foot, he sharply condemns the unnatural foot-
wear of his time and describes what he considers to be the most rational
shape of shoe.
If, then, we find in Camper efforts at comparative anatomy, this is only
evidence of his foresight, for as a general rule his contemporary zoologists
were content with purely superficial descriptions of types in the Linn^an
style. There were a few praiseworthy exceptions, however, among whom
John Hunter and Pallas deserve special mention.
John Hunter (172.8-93) was born in a country place in Scotland, the son
SEVENTEENTH AND EIGHTEENTH CENTURIES z6l
of a poor farmer. Being orphaned at an early age, he received a very indiffer-
ent education, a fact which influenced his whole life. He never even learnt
to spell his native language properly, nor at the beginning did he learn any
proper profession. At last, at the age of twenty, he went to his elder brother,
William, who had become a highly esteemed doctor in London and had been
commissioned to examine prospective army-surgeons. John began by assisting
at the dissection classes in connexion with this course, but during them he
taught himself anatomy, with such success that he was soon able to take
over the direction of the entire course. He continued to educate himself,
partly under his brother's and partly under other doctors' guidance, finally
receiving an appointment as surgeon, attached to the English fleet which
sailed to the Spanish main during the Seven Years' War. At the end of the
war he settled down as a physician in London, won a reputation as a clever
operator, and quickly obtained a remunerative post. He spent all his spare
time in anatomical and physiological studies, and as soon as his salary per-
mitted, he bought a house, in which he established a large anatomical mu-
seum. On this museum he sacrificed all that he could spare in the way of
time and money, so that at the time of his death it was undoubtedly the
finest of its kind in existence. He also gave private lectures in anatomy, but
he was not a particularly good lecturer. As a practitioner, on the other hand,
he was regarded as the best in London in his time. He was universally known
as an honest, benevolent, and charitable man, but his personal manners
showed his poor education, while his lack of self-control in particular gained
him many enemies. In a violent altercation with some of his colleagues he
got a stroke of apoplexy, which caused instant death. His museum was taken
over by the State and is to this day one of the sights of London in the sphere
of natural science. His manuscripts, however, were taken by a brother-in-
law, who first plagiarized them for his own benefit and then burnt them in
order to destroy all evidence of his plagiarism.
Hunter s work in comparative anatomy
Hunter's scientific work falls essentially within the sphere of practical
medicine; his theoretical researches were always intended as foundations on
which to base practical medical work. His famous museum was intended
for a similar purpose, but on the broadest lines; he collected all kinds of
animals, both higher and lower, dissected them, and experimented with
them, setting up the preparations that he made on anatomical principles.
Thus he applied for the first time in practice principles of comparative anat-
omy as a whole, thereby creating a precedent for future research of very
great value. Of his writings a treatise on the natural history and diseases
of the teeth has been of the utmost value to biology; in it he gives an account
of a systematic investigation into the origin and grov\th of the teeth that
is far in advance of any previous work of its kind. In a treatise on inflamma-
2.62. THE HISTORY OF BIOLOGY
tion of the blood and bullet-wounds he propounds a curious theory of the
blood as a vital principle, which he further developed in a number of lectures
on the musculature. He considers the blood to be a kind of primary matter
in the body, whence every other bodily substance is derived; all living matter
is of a similar nature, so that even the blood of one animal can be transferred
into another of a different genus (modern investigations show this to be an
error), and life is a kind of independent principle in the body which prevents
it from dissolving — a theory reminiscent of Stahl. Though Hunter pro-
duced a few solitary brilliant ideas, yet from a theoretical point of view he
did not contribute very much to the development of biology; his genius for
comparative anatomy was, however, probably greater than anyone else's in
his time, and in many respects it has borne fruit in more recent times.
Peter Simon Pallas was born in Berlin in the year 1741, the son of a
doctor, and studied medicine in his native country, at Gottingen, and at
Leyden. At the latter university he got his degree with an essay on intestinal
worms. He afterwards spent some years in Holland, working at zoological
collections from the tropics, which he described in a series of papers. In
1768 he was summoned by the Russian Government to take part in an im-
portant expedition which was being sent to Siberia to explore that country
from the point of view of natural history and economics. Pallas spent six
years travelling in Siberia, reaching as far as Amur, and he brought home an
immense quantity of scientific material, which he worked at in St. Peters-
burg for a number of years. In 1793 he was sent to explore the Crimean
district, which had just then become part of Russia, and he stayed there for
a long time, living on an estate which the Empress Catherine II gave him.
Finally, however, he moved back to Berlin in order to be in closer touch
with the scientific world, and there he died in 1811.
Pallas' s ivork on intestinal worms and on mammals
Pallas's contribution to the development of biology is particularly many-
sided. In his doctor's dissertation he incorporated all the observations he
was able to obtain dealing with intestinal worms and he sought to prove
that they enter the human body from outside — in his time it was univer-
sally assumed that they arose out of "tainted fluids" in the body. In a work
on the zoophytes he tries to find out the classification of these animals,
their conditions of life, and their relation to animals and plants. He endeav-
ours to prove that the zoophytes form a true transition between animals and
plants, following the ancient saying that nature never makes any jumps.
He also made a number of interesting observations, both anatomical and
biological, on worms and expressly points out how utterly heterogeneous
the Linnasan class bearing this name is. Primarily, however, Pallas is a
student of vertebrates. In his Spicilegia zpologica in particular — a collection
of monographs, issued in separate numbers — he describes in detail a number
SEVENTEENTH AND EIGHTEENTH CENTURIES x6}
of hitherto unknown higher animals, dealing with their anatomy, mor-
phology, and habits. Among his biological works, however, the place of
honour is held by his work on New Mammal Species from the Kodentia. In
this work he gives an account, with a thoroughness that was quite unprec-
edented, of the new rodents discovered by him in Russia and Siberia; in
it he endeavours to present not merely diagnoses as resulting from his ex-
aminations, such as his age was usually content with, but a true general
knowledge of the animals described, based on a close study of their exterior,
with careful measurements of every part of their body, thorough anatomical
investigations and illustrations, and detailed descriptions of the conditions
under which the animals lived. The anatomical section is particularly useful
and constitutes the best work so far carried out in the investigation of the
inner structure of the members of an entire order of animals. Though direct
points of comparison do not occur very often in the work, nevertheless the
descriptions are so detailed and at the same time so comprehensive that the
whole must be regarded as one of the really sound pieces of work that have
paved the way for modern comparative anatomy.
At this point we may close our account of the biology of the eighteenth
century. Before, however, proceeding to the cultural phenomena — already
hinted at above — that represent the basis of the natural science of the
nineteenth century, we must take a glance at a radical reform in another
sphere of natural science, which contributed towards the creation of modern
biology.
CHAPTER XII
THE FIRST BEGINNINGS OF MODERN CHEMISTRY AND ITS
INFLUENCE UPON THE DEVELOPMENT OF BIOLOGY
The phlogiston theory
So LONG AS chemical processes had their explanation in the phlogiston
theory, it was certainly possible to offer a provisional explanation of
a number of phenomena in the sphere of combustion and oxidization,
but any deeper insight into the material changes which both animate and
inanimate nature daily undergo was of course out of the question. In par-
ticular the qualitative side of the chemical process was, as far as this theory
went, inexplicable. In spite of this, the theory was stubbornly maintained
during the greater part of the eighteenth century, doubtless because so many
discoveries had been made under the assumption of its correctness, which
the chemists hesitated to interpret anew. For the rest a more accurate knowl-
edge of the process of combustion presupposed a knowledge of the types
of gas that play a part therein, and this knowledge was not acquired until
the latter half of the eighteenth century. The progress made in this field of
inquiry is primarily bound up with three names: the Englishmen Priestley
and Cavendish, and the Swede Scheele. Priestley deserves still further mention
as a discoverer in the biological sphere; Cavendish (173 1-1810) is best known
as the discoverer of hydrogen, and Scheele (i74z-86), one of the most bril-
liant experimental scientists of all time, succeeded in making, in spite of his
short life, a large number of chemical discoveries, his treatise On Air and Fire
becoming especially famous.
Joseph Priestley was born in 1733 of a Free Church family of the artisan
class living in the north of England. After studying in his sect's theological
training-college he was eventually ordained a minister and served in several
parishes, partly in Birmingham. An extreme radical, both in religion and
politics, he was a supporter of the French Revolution, which resulted in his
being subjected to personal persecution; the mob attacked him in his home,
which they pillaged, and he himself escaped with his life and fled to London.
As he found no peace there either, he emigrated to America and died there
in 1804. Priestley had begun to carry out chemical experiments independ-
ently; throughout his life he worked quite unsystematically, heating up
and treating with reagents everything that fell into his hands, but as he
possessed a great gift for arranging and observing his experiments, he did
x64
SEVENTEENTH AND EIGHTEENTH CENTURIES 165
some wonderful pioneering work. One of the chief results of his observations
was his discovery of oxygen, which he found by heating mercury monoxide;
further, he experimented successfully with carbonic acid, which brought
him into the sphere of vegetable and animal chemistry. He found that rats
kept in a volume of air that was confined by water died as a result of the
pollution of the air, but by letting green plants stand for a time in that
same air, it was so improved that fresh rats were again able to live in it
for some time. He found by a series of experiments that the air polluted by
the animals' breathing contains carbonic acid, or "fixed air," as he called
it. As a theorist Priestley was not particularly original; up to his death he
stubbornly maintained the phlogiston theory, which had already been aban-
doned by most chemists of his age.
The scientist who put the chemistry of combustion on the right road
was Antoine Laurent Lavoisier. He was born in Paris in 1743, the son
of a lawyer, and was given an excellent education, special attention being
paid to mathematics and natural science. He went in for an official career,
however, and in time became a "farmer-general" — that is, a titular mem-
ber of a body to which the French Government had leased the collection
of revenue. This system naturally gave rise to a good deal of abuse, and its
officials were not much better tolerated than the publicans of the Jews of
old. Lavoisier had never been guilty of fraud, but when the Revolutionary
tribunal condemned his colleagues, he was likewise involved in their fall.
Condemned for no reason at all, he was guillotined by the Terrorists in
1794. When his services to science were cited as grounds for mercy, the peti-
tion was met with the reply: "La Republique n a pas besoin de savants."
Lavoisier founds quantitative chemistry
It has been said of Lavoisier that he never discovered a new substance or
a new phenomenon, but that he introduced a new spirit into his science.
Even the system of weights and measures on which he based his reform had
been used before him by Hales and others, but Lavoisier was the first who,
in the study of chemical phenomena, consistently paid attention to the
weight conditions and in each chemical process determined their immuta-
bility, thereby making of chemistry an exact science in the same way as
physics. Thanks to Priestley's discovery of oxygen, he was able to account
for combustion and he gave the name of "oxgyen" to the gas which had
formerly been called " dephlogisticated air." Likewise, he established the
fact of water's being composed of oxygen and hydrogen, the latter discovered
by Cavendish. Moreover, he found out that heat is unweighable — a fact
which still further explained the process of combustion. He also applied his
weighing method to life-phenomena; he shut up animals in a confined volume
of air and by means of weighing determined the change brought about by
their breathing therein. He established the fact that oxygen is the component
■l66 the history of biology
of the air which is consumed by respiration and that it is substituted in the
lungs for carbonic acid. He saw chemical processes both in respiration and
in animal heat, as also in fermentation. His influence on the development of
science can scarcely be too highly estimated; through him chemistry was
led into entirely new channels; through the discovery that oxygen was a
constituent common to a mass of chemical elements, the latter could be
viewed from a common standpoint and could be given a nomenclature
which in part is still in use today. Moreover, to natural science in general
these discoveries meant a complete revolution, as they paved the way for
the knowledge of the indestructibility of matter. Lavoisier's association
with biology lies, of course, mostly in his knowledge of the respiratory
process. It was vegetable physiology in particular that felt the immediate
influence of the new advance in chemistry. Two examples of this are given
in the following.
Jan Ingenhousz was born at Breda, in Holland, in 1730, and studied
medicine at Leyden under Albinus. As a medical practitioner he was espe-
cially known for his skill in smallpox inoculation — an operation which in
those days was not unaccompanied by danger. Persons of high rank came to
him to be inoculated, and he was the recipient of distinguished and high
marks of appreciation. He died during a journey lo England in the year
1799. In the course of a previous visit to England Ingenhousz had learnt of
Priestley's above-mentioned attempts to "improve polluted air" by the in-
troduction of live plants, and he resolved to proceed with them in a more
extensive and systematic form. And in spite of the fact that his experimental
apparatus lacked variety and originality — he immersed different parts of
plants in water and collected the gas thus given off by them — he succeeded
in establishing a number of facts of fundamental importance for the knowl-
edge of plant life. He found that the production of " dephlogisticated air,"
which constitutes the plant's role as an air-purifier, is a prerogative of the
leaves, and particularly of their under side, and that it is brought about
exclusively by the influence of sunlight on the plant, whereas during the
night, and even in the shadow by day, a kind of air is produced that is
fatal to animal life, and that this air is produced by roots, flowers, and fruit,
while these latter, if enclosed in a confined air-space, render it impossible
for a light to burn in it. Ingenhousz also carried out quantitative investi-
gations, though of a somewhat primitive nature. In this field both he and
every one of his contemporaries were far outrivalled by a younger scientist,
who, it is true, had the inestimable advantage of being able to avail himself
of Lavoisier's new methods.
Nicolas Theodore de Saussure was born at Geneva in 1767. His father
was 3 scientist of repute and was interested in botany, but his real mefier
was geology. The son also eventually became professor of geology, after-
SEVENTEENTH AND EIGHTEENTH CENTURIES %^-J
wards entering the Genevan representative council and making a great rep-
utation both as a scientist and as a public citizen. He died in 1845. He was
both a chemist and a physicist, but he is chiefly known for his work on
vegetable physiology, spending years in the investigation of the subject and
finally publishing his results in 1804. The greatest service performed by this
work lies in the fact that here for the first time the quantitative method of
chemical research, as founded by Lavoisier, together with its results, were
systematically applied to living subjects of investigation. This opened up
for Saussure entirely new possibilities for methodically organizing his experi-
ments that his predecessors never possessed. He enclosed plants and parts of
plants in a quantity of air which had been previously weighed and carefully
analysed, and after having let them live there under different conditions,
in light and in darkness, he investigated the changes in the composition of
the air which their manifestations of life had brought about. He thus estab-
lished the quantitative relation between the amount of carbonic acid ab-
sorbed by the plant in light and the quantity of oxygen simultaneously
given off by it. In the same way he found out the quantity of oxygen absorbed
by a plant at night, and also the quantity of water consumed in association
with the absorption of carbonic acid that is required for the growth of the
plant. While the plant was thus found to derive the quantitatively most
considerable portion of its nourishment from the air, Saussure on the other
hand established the indispensability of the mineral constituents which it
drew from the earth, and which he determined by careful analyses of the
ashes of the plants investigated. Finally, he also found out that the per-
centage of nitrogen that the plants possess is primarily absorbed in the form
of ammoniac associations. On the other hand, Saussure was wrong in think-
ing, in contrast to Ingenhousz, that the green colour of the leaves is not
essential to their vitality — a misconception (based on the existence of red
leaves in certain varieties) that, owing to his authority, was long associated
with that line of research.
But while Lavoisier's new method was thus immediately applied to
biology with a large measure of success, the more speculatively inclined
scientists were led by it to make bold guesses — as is usually so with new
discoveries. In the romantic natural philosophy we shall find ideas which
were awakened to life by the great revolution in chemistry.
CHAPTER XIII
CRITICAL PHILOSOPHY AND ROMANTIC CONCEPTIONS
OF NATURE
I. Kant and his Immediate Successors
Mafer/alism and spiritualism i)i the eighteenth century
THE TRANSITION PERIOD between the eighteenth century and the suc-
ceeding era is characterized by the violent political and social
convulsions beginning with the French Revolution in 1789 and
ending with the fall of Napoleon in 181 5. During this period came into being
the modern social system, which, based on the claim of the private citizen
to be allowed both to determine his own actions and to take part in the
administration of the State, is sharply contrasted with that of the preceding
age, with the State possessing unlimited authority in all matters, both secu-
lar and spiritual. But even from a purely scientific point of view the begin-
ning of the nineteenth century involved a radical revolution, which had
been long preparing, like the political revolution, throughout the centuries.
In the eighteenth century's conceptions of nature and life, the two tendencies
described in the foregoing — the mechanical and the mystical-spiritualis-
tic — appear in deep contrast to one another. Out of the former, which has
its origin in the natural philosophy and natural-scientific research of the
seventeenth century, and which, like its predecessors, seeks to explain
natural phenomena on purely mechanical lines, there develops towards the
close of the eighteenth century — during the so-called "Era of Enlighten-
ment" — a general materialism of the kind that we have seen in La Mettrie:
a conception of life expressing itself partly in a dogmatically formulated
theory of existence as a play of exclusively material forces, and partly, in
the ethical sphere, in a doctrine of a state of blessedness common to all
mankind, based ultimately on the liberty to enjoy life independent of tradi-
tional rules of conduct. This doctrine, which assumed its best-known and
most popular form in Holbach's work Systeme de la nature, is remarkable for
its readiness to answer every conceivable question in accordance with the
formula, laid down once and for all, that, provided the mechanical explana-
tion of nature is maintained, the most daring constructions of thought and
the weakest verbal subtleties may pass as complete scientific evidence.
Intellectual superficiality and banal hedonistic morality thus became marks
i68
ID
2
"§iB
H
W
O
pq
CO
W
Pi
<
SEVENTEENTH AND EIGHTEENTH CENTURIES T.G<^
of the enlightened philosophy and contributed in succeeding generations
towards concealing its services in the political and the social sphere; the
philosophers of enlightenment have striven unceasingly for humanity and
tolerance in the life of the State, and in that respect their activities have
left a deep impression on the social life of our own day. Parallel with the
philosophy of enlightenment, however, there developed another, entirely
contrasted, conception of nature, the precursors of which had been Paracel-
sus and van Helmont, and which, possessing in Stahl, Swedenborg, and
Caspar Friedrich Wolff its scientifically most important representatives,
appears throughout the eighteenth century under various forms; a view of
life which sees in natural phenomena an expression for the operations of
spiritual powers, whereas, according to its tenets, the mechanical explana-
tion of nature admits of only a superficial observation of what takes place,
without any insight into that inherent connexion in existence which the
spiritual powers imply. This attempt to regard nature as a living entity, to
look for connexions in phenomena where, when viewed superficially, none
are apparent, has constituted this tendency's greatest service, besides which
the freedom of mechanical principles, in many cases, admitted of greater
liberty in the interpretation of special phenomena, as Wolff's embryological
and Sprengel's botanical investigations proved. The weakness of this spirit-
ualistic view of nature has lain in the frequent desire to solve by mystical
formulas problems the solution of which would have required observation
and deep thought, and, generally speaking, in its tendency to degenerate
into meaningless phrases. As, moreover, this natural mysticism was asso-
ciated with moral and religious speculations and was upheld by specially
founded mystic communities, there was thereby created that extremely un-
sound "secret wisdom" that under various names and forms spread with
incredible rapidity at the close of the eighteenth century, in spite of protests
and ridicule on the part of the adherents of enlightenment.
Kant and his -philosophy
Besides these two directions of thought, which offered, at least in their
more extreme forms, but slender possibilities for the further advancement
of science, there appears towards the close of the eighteenth century a new
system of thought which really gave the scientific activities of the next
century their peculiar character — namely, critical philosophy. Its founder
was Emmanuel Kant (1714-1804), whose life's work has undoubtedly rep-
resented the greatest contribution to the history of human thought since
Socrates and Plato, and for this reason his work merits attention even as
concerning the history of biology. Kant was born, lived, and died at Konigs-
berg, in Prussia, where he was professor of philosophy and applied himself
entirely to his work as a thinker and teacher. In his youth he had studied,
besides philosophy, certain exact sciences, chiefly physics and mathematics,
■L-JO THE HISTORY OF BIOLOGY
and throughout his life he retained his interest in natural research, not
least in biology. His first papers, in fact, dealt with mechanical and cos-
mological problems; the best known of these is his Allgememe Naturge-
schkhte und Theorie des Himmels, in which he tried to set up a mechanical
theory concerning the origin of the universe. On this subject Swedenborg
and BufFon had been his precursors; BufFon, an account of whose cosmological
theory has been given above, seems to have been his chief source of inspira-
tion. In contrast to the latter, Kant believes that the planetary system has
evolved from a collection of dust particles which moved in space and even-
tually became concentrated. This theory, the details of which need not be
recounted here, all the more so as it has often been referred to, testifies to
Kant's efforts to find a mechanical explanation of existence. Towards the
end of the work, however, he becomes involved, doubtless under the in-
fluence of Swedenborg, in fantastic speculations about life on other heavenly
bodies; he believes that on the more distant planets, Jupiter and Saturn,
there are beings of a higher order of intelligence than that of man — this
because the inhabitants of the more distant planets must be made of lighter
material in order that the less intense solar heat there may set them in mo-
tion; but the lighter the corporeal matter the greater the intelligence, while
heavy bodily fibres and dense, "sluggishly cooking" fluids result in in-
ferior abilities. Strangest of all, he cites Newton's calculations in support
of this theory, which might more naturally have originated from the earliest
Greek philosophers. Kant, however, soon rose above these fantasies; in a
paper published ten years later entitled Trdume eines Geistersehers he settles
with Swedenborg, as indeed with all metaphysical speculations upon the
relation between spirit and matter. He ironically examines all the old theories
about the location of the soul — now existing everywhere in the body, now
located in a small section of the brain — and finally proves the impossibility
of determining how the soul influences the body or whether spiritual beings
can exist without material space; reason is as little able to decide this ques-
tion as it is to determine how anything can be a cause or can possess a force —
which are all matters that can only be determined by experience; and alleged
experiences of single individuals, such as Swedenborg's visions, cannot form
the basis of a law of experience for the very reason that they are isolated
cases. He ends by pointing out that there certainly are many things that we
do not understand, but there is also a very great deal that we do not need
to understand. We must be quite clear as to what is necessary for us to know
and what in that respect can and must be dispensed with.
Kant, having thus exposed the futility of the old metaphysical specula-
tions, spent more than ten years in trying to find out the limitations and
conditions of the human capacity for knowledge in general. The result of
these researches he recorded in his Kritik der reinen Vernunjt, published in
SEVENTEENTH AND EIGHTEENTH CENTURIES zyi
1781 — one of the most epoch-making works in the whole history of human
development. Kant's purpose — which, in fact, he at least partly achieved
^ — is to lay the foundations of a new philosophy, to meet not only all the
needs of human life in the way of knowledge, but also its moral and religious
aims. The many different points of view from which he examines the work-
ings of the human mind, as well as the laws he lays down therefor, cannot
of course be recounted here. Chief in importance for the future advancement
of natural science is his attempt to determine what justification natural
science has for assuming the truth of the knowledge of nature which it
expounds. Kant first of all discusses the ideas of space and time and finds
that they are not grounded in experience, but in human nature itself; all
experience, on the contrary, is based on our having the ideas of space and
time that we have. And the same part that time and space play in our views,
the idea of causes plays in our understanding. The knowledge we gain by
experience is a knowledge of the phenomena that appear to us owing to
our organization's being what it is. What those things that cause the phe-
nomena are like in themselves we can never know for certain. Natural
science is thus a knowledge of reality such as we observe it, not a knowledge
of reality as it actually is. Natural laws are based on our own capacity for
knowledge and are binding on us because this capacity has certain funda-
mental qualities that are the same for all men. Natural science is thus fully
justified in drawing its conclusions in the world of experience; on the other
hand, it can never give any enlightenment as to the intrinsic meaning of
things — that is, what is not phenomenon — nor indeed does it need to do
so for the purpose of its physical explanations; but even if, say, some influ-
ences from the immaterial world were to arise, it should pass them over
and base its explanations upon what the senses are able to reveal and what
is reconcilable in accordance with the laws of experience, with our actual
observations. On the other hand, all things on which the experience of the
senses can give us no knowledge, such as what the soul, the world, God,
actually are in themselves, fall outside any rational knowledge. Of these
things, then, we can know nothing — we can maintain neither their exist-
ence nor their non-existence. But for that very reason we are able, if our
feelings require it, to take them for granted; we are justified in believing in
God, in the immortality of the soul, and in the free will, and reason has
no right to reject any such belief as irrational. These things are, in fact, a
part of practical reason — that sense of duty and right which Kant is firmly
convinced is inherent in everyone; that which says, not ivhy we act in this
or in that way, but hotv we should act in order to obey the dictates of con-
science within us. — Kant himself, in spite of his keen criticism of the
life of the human soul, was an ideally minded personality throughout —
an enthusiast over such questions as human justice and social equality, who
xyz THE HISTORY OF BIOLOGY
hoped for universal peace in the future. The highest feeling he knew he used
to express in the following words: "The starry heavens above me, the sense
of duty within me." These words are actually carved on his gravestone.
His influence
Through his critical philosophy Kant has played an important part in
human cultural development in general, and not least in scientific develop-
ment. Thanks to his criticism, biology was freed from the question, which
had so often arisen and yet had never been solved, of the relation between
soul and body; biological research had, as its exclusive mission, to explain
the material course of the phenomena of life, while the investigation of the
spiritual side of the soul-life became the function of the science of psy-
chology, employing entirely different methods. But in other respects, too,
Kant's critical philosophy exercised an influence upon the development of
biology in the century that followed; many of its leading biologists have
been keen supporters of Kant; for instance, Johannes Miiller, to mention only
one of the most eminent. But Kant's practical criticism of reason has also
indirectly affected the development of natural science. He thereby established
that reason can neither prove nor disprove man's personal ideas of faith and
conscience, so that any attempt to influence what the individual holds in
high esteem and deep reverence is both unjustifiable and irrational, whether
it is done in the name of the Church or in that of science. His principle,
just and reasonable though it is, has nevertheless found it difficult to gain
a hearing; ever since then, and indeed up to the present day, there have been
disputes between "faith and knowledge," brought about not least by the
fact that the ecclesiastical authorities claim that their doctrines shall be
accepted in their entirety as objectively true. The Roman Catholic Church
in particular has banned Kant and his philosophy. But even his contem-
poraries found it difficult to reconcile themselves to the strict self-control
that Kant enjoins upon human thought; that nothing was to be known of
"things in themselves" annoyed both the old philosophers of enlighten-
ment, who found Kant's thoughts difficult to grasp and oversubtle, and also
the champions of the mystical-romantic class, which strove after a uniform,
comprehensive view of existence. In particular, thinkers in the latter di-
rection, while adopting certain of Kant's principles, thought to bring human
knowledge beyond the contrast between personal consciousness and the
''Ding an sich" ; as a matter of fact, the whole of the beginning of the nine-
teenth century was full of efforts of this kind, which have left their mark
on every science, and indeed on the whole of human culture during this
period of history. And so far as they influenced the development of biology,
they will be briefly touched upon in the following pages.
JoHANN Gottfried Herder (1744-1803) was a fellow-countryman and
disciple of Kant's. He was ordained a minister and held a living for a time
SEVENTEENTH AND EIGHTEENTH CENTURIES 173
in Riga, afterwards spending some years travelling in Europe, and finally,
on Goethe's recommendation, becoming court chaplain at Weimar. At the
same time enthusiastic and irascible, he had great difficulty in getting on
with people; at times even he and Goethe would be on bad terms, but they
would soon become reconciled again. As a poet and student of folk-lore
Herder has contributed much to literary history. Pronounced romanticist
as he was, he sought earnestly for a uniform conception of existence; in these
efforts Spinoza was his principal master — he rescued the latter's writings
from oblivion and was an ardent supporter of the more mystical views con-
tained therein, whereas Kant's criticism attracted him but little. In his prin-
cipal work, Ideen xur Fhilosophie der Geschkhte der Menschheit, Herder tries to
prove how one and the same spirit dominates the whole of nature; all living
beings have been created -according to one common plan; their various char-
acteristics correspond to their peculiar functions in life, which finally reaches
full perfection in man. In the whole of this conception of the course of life
Herder is a precursor of the romantic philosophy, which left such a deep
impression even on biological history.
JoHANN Gottlieb Fichte (1761-18 14) is generally regarded as the first
of the purely romantic philosophers. The son of poor parents, he suffered
many hardships before becoming a professor, first at Jena, where, on account
of his strictly moral principles, he came into conflict with both professors
and students and was finally dismissed for "atheism"; and then in Berlin,
where he worked hard for the elevation of morals and of the national spirit
under the oppression of Napoleon's rule. His philosophy, too, is mostly
concerned with ethics; he is of only indirect importance in biological history,
as having been the teacher of Scheiling, the founder of natural philosophy.
Fichte bases his philosophical speculation on Kant, but he also felt the in-
fluence of Spinoza. Kant thought that our consciousness gives us the idea
that we have of a thing, whereas the thing itself is unknown to us. Fichte
also starts from the idea of consciousness, but denies the existence of the
thing in itself: he believes that the consciousness or the ego, ''das Ich," as
he calls it, is the only true thing existing; through its operation it then gives
rise to existence apart from itself — "the ego places the non-ego," runs
the oft-quoted phrase, which primarily refers to the creative work which
the moral will of man performs, for the moral will is man's true ego and the
central point in the whole of Fichte's extremely abstract and involved spec-
ulations. But besides the individual ego, Fichte assumes an "absolute ego"
— a kind of world-soul, which can be attained by man only through "in-
tellectual intuition" — a kind of mystical impulse on the model of Spinoza.
It was Scheiling who further developed his idea, making it one of the foun-
dations of his natural philosophy.
Friedrich Wilhelm Joseph Schelling was born at Leonberg in
2.74 THE HISTORY OF BIOLOGY
Wiirttemberg in 1775. He was the son of a clergyman and from childhood was
destined for the same calling. He developed early; after a brilliant career at
school he matriculated at the age of fifteen and in his twentieth year had taken
his doctor's degree in both philosophy and theology. His first philosophi-
cal studies had dealt with Spinoza, Kant, and Fichte. He afterwards spent a
couple of years as a private tutor at Leipzig and there studied natural science,
chiefly chemistry and physics. At the age of twenty-two he published a
work which at once brought him fame — Ideen zu einer Philosofhie der Natur.
Thanks to this work, he was appointed assistant professor at Jena — Goethe,
who was much interested in the book, had recommended him to the Saxe-
Weimar Government for the post — and there he came in contact with a
circle of men and women of genius with pronounced romantic views on
science and art. There was one who exercised special influence on him —
Caroline Michaelis, a gifted and energetic woman, who, although she was
twelve years older than he and had had a somewhat adventurous past,
became his wife and highly influenced his work as an author. After her
death, in 1809, Schelling's influence actually came to an end. Six years be-
fore, however, he had already left Jena, where, through his extraordinary
insolence, he had acquired many enemies — with one or two of these he
entered into a dispute that ended by their all being condemned for libel.
After this he was for a time professor at Wiirzburg. He then spent a long
time in Munich as secretary to the Academy, but he was finally summoned
to Berlin (in 1841) for the purpose of using his romantic philosophy to
counteract the increasing radicalism. In spite of the support of the Govern-
ment, however, he was utterly defeated; his enemies published his lectures
with insolent comments, with the result that he withdrew altogether from
public life. He died in 1854. His character was conspicuous for ostentatious
egotism and uncurbed violence by the side of a devoted faith in the doctrines
he expounded. Early successes spoiled him, and when later he was confronted
by opponents who did not allow themselves to be frightened by his over-
bearing manners and scornful polemics, his creative power vanished entirely.
The work that brought him fame was completed before his thirtieth year;
the half-century that he lived after that added nothing to his renown.
As a thinker Schelling based his ideas on Spinoza and Fichte. With Kant,
on the other hand, he had but little sympathy; the doctrine of the strict
limitation of the capacity of human reason that Kant taught was really the
direct opposite of what Schelling desired and thought himself able to pro-
duce. His relations with Fichte, however, were at first those of a loyal pupil,
though he later broke entirely with him. It was from this master of his that
Schelling borrowed the principle of the ego as the basis of everything, both
in the spiritual and in the material world. The greatest influence on Schelling,
however, was exercised by Spinoza, with his doctrine of spirit and matter
SEVENTEENTH AND EIGHTEENTH CENTURIES 175
as different forms of one and the same "substance" and with his principle,
derived therefrom, of the validity of the laws of human reason even in nature.
When afterwards, in Leipzig, Schelling became acquainted with natural sci-
ence, chiefly with chemistry, which was then making great strides, there
awakened in him a desire to create, like Spinoza, one common system of
thought embracing the whole of existence, which was to prove the connexion
between the worlds of nature and of the spirit, in that the world of nature
would be derived from that of the spirit, and, vice versa, the world of the
spirit from that of nature. The latter became the aim of Schelling's natural
philosophy proper, which in one place he terms " Spinozismus der Physik."
Schelling s natural philosophy
From this a new natural science was to arise, which was not only to observe
individual phenomena and from them derive certain universal principles, but
which would actually understand the fundamental forces that cause all that
happens in nature. Thus it was a program of natural research directly opposed
to that developed theoretically by Bacon and practically by Galileo, which,
indeed, research has followed since then. Nevertheless, Schelling expresses
the deepest contempt for this natural research; in one place he calls Bacon,
Newton, and Boyle the bane of natural science, and Lavoisier's chemistry
is treated with no less disdain. It is natural enough that the so-called Spinoz-
ism which Schelling would put in its place should become a mere dogmatic
system of thought; moreover, as he was entirely lacking in patience and
consistency in matters of detail, his theory became vague and fragmentary.
In view of the great influence it exercised on the development of biology,
however, an attempt must be made to describe it.
In a paper entitled Darstellung ineines Systems, which Schelling, after the
manner of Spinoza, wrote in the form of a series of statements and proofs —
though unfortunately entirely without that strictly binding logic which
characterizes every sentence of the great Jewish thinker — he describes first
of all the ' ' absolute reason " as " eine tot ale Indifferent, des Subjektiven und
Objektiven," which is to be attained by thinking of reason while being fully
abstracted from one's thinking self. This is indeed ultimately the same as
the mystical view with which Spinoza concludes and with which Schelling
thus, strikingly enough, begins. Outside this reason there is nothing, and
in it is everything. The supreme law governing the existence of reason is
the law of identity — that is, A = A. "Die absolute Identitat kann nicht un-
endlich sich selbst erkennen, ohne sich als Subjekt und Objekt unendlich xu setzen.
Dieser Satz. ist durch sich selbst klar." Thus arises the contrast between subject
and object, by which Kant, as we know, meant the consciousness that con-
ceives and the thing which is conceived, and which in Schelling means about
the same. Further on, the absolute identity is said to correspond to the uni-
verse, whereupon the subject and object are denoted by A = B; finally
2.76 THE HISTORY OF BIOLOGY
+
matter corresponds to A = B, because in matter the objective predominates.
Then the absolute indentity is identified with light, the opportunity being
here taken to sneer at Newton for his spectral investigation and to compli-
ment Goethe upon his optical theories. From matter, on the other hand,
are derived gravity, cohesion, and magnetism: "Die Materte im ganxen ist
ah ein unendlkher Nlagnet an%usehen," and " Der Magnetismus ist bedingend der
Gestaltung,'' with the result that " alle Korper sind blosse Metamorphosen des
Eisens." Electricity and magnetism are indentified according to a deriva-
tion that reasons of space compel us to pass over, as also the derivation of
heat from the foregoing. Finally, organism is derived from the absolute
identity. As an example of Schelling's biological speculation may be cited
one of the paragraphs in the work in extenso (the spacing is Schelling's):
,, Der poten^ierteste positive Pol der Erde ist das Gehirn der
Tiere, und unter diesen des Nlenschen . Denn da das Gesetz der Meta-
morphose nicht nur in Ansehung des Gan^en der Organisation, sondem auch in
Ansehung der einxelnen gilt, das Tier aber der positive (^Stickstoff^ Pol der allge-
meinen Metamofphose ist, so wird im Tier selbst ivieder das hbchste Produkt der
Metamorphose der vollkommenste , d.h. potenzjerteste Pol sein. Nun ist aber (wie
bekanni^ das Gehirn das hochste Produkt u.s.tv. Also etc.
, , Anmerkung i. Der Beweis dieses Sat^es ist jreilich nicht aus den chemischen
Analysen %u jiihren, aus Grunden, ivelche kunjtig allgemein werden eingesehen
werden. . , .
,, Anmerkung 2. Das Bestreben der Metamorphose im Tierreich geht, wie aus
dem bisherigen leicht xu schliessen ist, notxuendig durchgangig auf die reinste und
potenzierteste Darstellung des Stickstojfs. — Dieses geschieht in dem gebildeten
Tier fortivdhrend durch den Process der Assimilation, der Kespiration, tuelche
bloss daxu dient, den K.ohlenstojf vom Blut losxureissen; ruhiger und nicht mehr
in einem stetigen tinunterbrochenen Process, gleichsam als ob die Natur iiber sich
schon zu Kuhe gekommen ware, durch die sogenannte ivillkurliche Beivegung. —
Das erste ruhende Tier stellt die bereits ganz aus sich selbst herausgekommene
Erde dar; mit der vollkommensten Gehirn- und Nervenmasse aber ist ihr Imierstes
entfaltet und das Reinste, das die Erde der Sontie gleichsam als Opfer darbringen
kann.
,,Zusatz J. Das Geschlecht ist die Wurzel des Tieres. Die Blute das Gehirn
der Pflanzen.
,, Zusatz 2.. Wie die Pflanze in der Blute sich schliesst, so die ganze Erde im
Gehirn des Menschen, welches die hochste Blute der ganzen organischen Metamor-
phose ist.^'^
This quotation may suffice. If the reader desires more he is referred to the
original, the 159 paragraphs of which are, some slightly less, others some-
what more, absurd than the one quoted. Quite out of our subject is Schel-
SEVENTEENTH AND EIGHTEENTH CENTURIES xyj
ling's transcendental philosophy, which resolves itself into a glorification
of art, as being the identity of the conscious and the unconscious, and in
which subject he certainly felt far more at home than in natural science.
The influence of his system
But even in regard to natural philosophy it would be entirely unhistorical
to dismiss Schelling as simply and solely a half-witted fool, as is so often
done. This is at once inadmissible, owing to the extraordinary influence he
exercised on his own age. And it must be admitted that among all the identi-
fications and derivations that constitute his system, there are, besides much
madness, a number of really brilliant ideas, which, although expressed as
mere fancies, nevertheless undoubtedly exerted an influence upon the future
development of science. Thus we may at least suppose that Schelling's com-
parison of electricity and magnetism was not without its influence upon
Orsted, who along experimental lines discovered electromagnetism and who
in his youth was a great admirer of Schelling. It should also be noted that
Schelling had a keen eye for the physiological contrast between plants and
animals, which lies in the former's oxygen-production and the latter's
oxygen-resorption; the significance of this contrast for the general economy
of nature he has realized and expressed quite clearly, though, it is true, he
draws the odd conclusion that the plant has no life, for its arises merely
through the development of the life principle and possesses only the sem-
blance of life "im l^ioment dieses negativen Processes." — The whole of the
extraordinary thought-system which he built up finds its explanation partly
in the vast possibilities which the new gas-chemistry had just then opened
up for research and speculation — even in our own time hopes of the solu-
tion of the riddle of life have more than once been placed upon important
discoveries in the field of natural science — partly in the change from criti-
cism to dogmatic philosophy which Fichte had brought about with his
theory of the ego as the origin of all things and which was in complete har-
mony with the romantic tone that was peculiar to this epoch, particularly in
Germany. People dreamt of a uniform conception of existence, they looked
for spiritual forces in nature, they had grown accustomed to the mystical
dreams that were propagated by a number of secret brotherhoods, and all
these vain strivings Schelling met with his explanation of existence as an
"absolute identity," an explanation that was no more dogmatic, indeed, but
certainly more poetic, than La Mettrie's and his successors' materialism,
which had constituted the natural philosophy of the previous genera-
tion. What, after all, makes Schelling's natural philosophy useless from the
point of view of natural science is its absolute lack of practical value; if the
object of natural science is to extend and consolidate man's dominion over
nature — and that has indeed been its aim ever since the days of Aristotle
and Hippocrates — then certainly most of Schelling's efforts have been in
xyS THE HISTORY OF BIOLOGY
vain, however much genius he put into his work. But even as a purely specu-
lative thought-system it suffers from serious defects — inconsistency, daring
conclusion, and lack of cohesion. All this, however, Schelling took quite
lightly; he was indeed a genius and an artist, and these factors work, accord-
ing to his theory, half unconsciously and without being worried by the
pedantry of the man in the street. It was these very faults, however, that soon
proved his undoing; it was just in the purely theoretical sphere that his
philosophy was out-distanced by another system, the Hegelian, which was
equally abstract and unreal, but far more consistently thought out, and be-
sides, from the scientific point of view, it had the undeniable advantage of
not involving nature in its speculations.
Georg WiLHELM Friedrich Hegel (1770-1831) was a fellow-country-
man and school-friend of Schelling's and, though older, was at first under the
influence of his precocious friend. However, he eventually worked out for
himself a theory of his own and in his first independent work published a
severe criticism of Schelling's theory of the absolute, which is described as
the "simplicity of the emptiness of knowledge; a night in which all the cows
are black." Hegel eventually became professor in Berlin and founded a well-
attended school, which he subjected to strong discipline. What impressed
his pupils, and indeed his entire age, was, besides his commanding person-
ality, the splendid consistency that he developed in his system of thought.
His dialectical method, however, according to which every idea has its
opposite, both of which are afterwards brought together and combined into
one larger idea, has no concern with our subject, especially as Hegel and his
disciples expressed the deepest contempt for nature and its study — which
gradually resulted in natural scientists' turning the tables by generally re-
garding all that is meant by philosophy as empty prattle about empty
fancies. On the other hand, Hegel performed a great service to the study of
history in insisting upon the necessity of ascertaining not only the events
that took place, but also the spiritual movements that brought them about.
A similar position to that of Hegel in Germany was held in Upsala by Kris-
TOFER Jakob Bostrom (1797-1866), who for half a century governed that
university and made of Linnasus's ancient seat of learning a centre of abstract
speculation.
But though Schelling was thus worsted in the theoretical sphere, his
natural philosophy survived as a general theory of life with the support of a
whole generation of contemporary scientists. The cause of this strange phe-
nomenon must partly be sought in the fact that there was no other equally
comprehensive explanation of nature available, and some such explanation
was an absolute essential of existence at that time. But there were many con-
tributing causes thereto, including the fact that Schelling's natural philoso-
phy was embraced by a man who was regarded by his age as an authority in
SEVENTEENTH AND EIGHTEENTH CENTURIES 179
every sphere of culture — namely, Goethe, the poet and universal genius.
As is well known, he has been an influence even in the field of biology and
it is this side of his work that will be described in the following section.
z. Goethe
JoHANN Wolfgang Goethe was born in 1749 of wealthy middle-class parents
at Frankfurt am Main. He studied jurisprudence, first at Leipzig, then at
Strassburg, and after passing his law examinations practised for a time as
a lawyer and at the same time acquired a reputation as a poet. In 1775 ^^
came to the court at Weimar, which was interested in literature, and there,
thanks to his brilliant intellectual and personal advantages, obtained an
eminent position — not only as poet and organizer of the pleasures of the
court, but also as an official he held the highest appointment in the little
Saxon capital. For a long time he held the reins of government with success
as Minister of State to the principality of Saxe- Weimar. In 1786 he made a
journey to Italy, which lasted two years and which proved of decisive im-
portance in his life, especially in regard to his scientific work. Having re-
turned home, he gradually withdrew from public life and devoted himself
whole-heartedly to poetry and science. Active and possessing his full in-
tellectual powers to the last, he attained a great age, dying in i8t,x.
Even as a child Goethe had evinced a lively interest in nature; he ex-
amined flowers and carried out experiments in electricity and magnetism.
In his poems, too, there was conspicuous from the very first a keen interest
in nature — a gift of observing and describing its life in its different phases,
which greatly contributed to his fame. During his student days he received
varied impressions from the extremely chequered intellectual life prevailing
in Germany at that time; he became acquainted with French materialism,
which seemed to him dry and unanimated; on the other hand, he engrossed
himself in mystical literature, studying the writings of Paracelsus, van Hel-
mont, and Swedenborg, which made a somewhat deep impression on him
and influenced his poetry. In Strassburg he made the acquaintance of Herder
and, as he himself declares, his association with him increased his inter-
est for the study both of nature and of human development. Like Herder,
Goethe admired Spinoza and sought in him a basis for the unity between
spirit and nature that he desired to find in life.
Goethe' s anatomical researches
At Weimar Goethe's interest in the natural sciences was increased through
his intercourse with scientists at the University of Jena and through periodi-
cal collaboration with Herder. While the latter was putting the finishing
touches to his above-mentioned Idcen, Goethe was studying anatomy ac
z8o THE HISTORY OF BIOLOGY
Jena. Herder, it will be remembered, was endeavouring to find one com-
mon type for the form and functions of human beings and animals. During
the latter half of the eighteenth century, moreover, a dispute had been
going on with regard to the relation of man to the apes, about which we
heard through La Mettrie and Camper; the former, indeed, had sought to
prove that the orang-utan was a kind of human being, which it should be
possible to civilize, and the indignation the theory aroused in the Chris-
tian-conservative people had found expression in violent polemics, while
Camper, through his paper on the anatomy of the orang-utan, referred to
above, gave support to those who maintained the dignity of man. Goethe,
who as a young man was somewhat averse to religion, to which his fa-
mous poem Prometheus in particular testifies, entered into the dispute on the
side of the materialists. Camper had asserted that in the facial skeleton of
the orang-utan there is a suture which, starting from the nasal cavity, ex-
tends on either side as far as the space between the corner tooth and the fore-
most front tooth; this suture does not exist in man, in contrast to the apes
and other mammals. In consequence of this Goethe wrote a short treatise in
which he maintains that the intermaxillary bone, which terminates in the
said suture, is found also in man — an assertion based chiefly on the exist-
ence of sutures which in the gum and above it separate the bone in question
from the upper jaw and adjoining bones. Goethe also described it as existing
in certain other mammals in which this bone had not previously been found.
The treatise was sent in 1784 to Camper, who expressed courteous thanks for
it and specially complimented Goethe on having established the existence of
the bone in the walrus. In regard to the discovery in man, on the other hand.
Camper had no remarks to offer, and that for sound reasons; as a matter of
fact, the bone had been known ever since Vesalius's days and had been de-
scribed in man, in whom in the embryonic stage it is clearly separated, while
in full-grown individuals its outer suture disappears. This difference between
man and the ape existed just as Camper had pointed out, and for obvious
reasons Goethe had not been able to disprove it. The fact that he imagined
he had "discovered" the intermaxillary bone in man was no doubt due to
the accident that some text-books of that time treated the incompletely
separated bone in the full-grown man as if it were one with the maxilla
superior. The claim to this discovery has on Goethe's authority even reap-
peared in literary histories and is believed by the public, unjustifiable though
it is. Goethe's pamphlet on the question remained for the time unprinted,
presumably owing to lack of encouragement on the part of the specialists;
at any rate, there were no financial obstacles standing in the way of its
publication.
Goethe, however, continued his anatomical experiments and finally
published the theoretical views at which he arrived, in a paper entitled
SEVENTEENTH AND EIGHTEENTH CENTURIES x8l
Erster Enfwurf einer allgemeinen Einleitung in die vergleichende Anatomic, dated
1795. He starts with the principle that natural science is on the whole based
on comparison — a principle upon which Aristotle had already laid great
emphasis. As the standard of comparison Goethe sets up an ideal type, with
which the anatomical details in each animal form are to be compared. Thus
one should at once be able to interpret an anatomical detail in an individual
by comparison with the ideal type. Goethe offers no detailed description of
what he imagines this type to be like, and indeed it would have been difficult
to conceive one. Herder's above-mentioned speculations on an ideal type have
obviously influenced Goethe here far more than the already existing com-
parative anatomy such as Buffon, Daubenton, and Camper practised. Even
in this theory Goethe indulges in wild philosophical fancies, as when he
states that the tail of mammals ''ah eine Andeutung der Unendlichkeit organischer
Existenzen angesehen iverden kann,'' or when he says of the body of the snake
that it is " gleichsam unendlirh," because it does not need to expend matter
and force on extremities. This paper also remained in manuscript form for
the time being.
His metamorphosis of plants
Before this, however, Goethe had published the treatise that is generally
acknowledged to be his principal contribution in the field of natural science
— namely, his Versuch, die Metamorphose der Pflan^en xu erklaren, which was
printed in 1790. The gist of it is, briefly, that the leaves of plants gradually
develop through "metamorphosis," which first gives rise to cotyledons,
then to the stem-leaves, and finally to the flower-leaves: food-leaves and
petals, stamen and pistils. This metamorphosis can partly be "regular" or
' 'progressive, " " welche sich von den ersten Samenblattem bis zur letzten Ausbildung
der Frucht immer stufeniveise wirksam bemerken Idsst, und durch Umwandlung einer
Gestalt in die andere, gleichsam auf einer geistigen Leiter xu jenem Gipjel der Natur,
der Fortpflanzimg durch z^vei Geschlechter hinaujsteigt.'" Irregular metamorphosis
is one of nature's retrograde steps: "ivie sie dort mit unwiderstehlichem Trieb
und krdjtiger Anstrengung die Blumen bildet und xu den Werken der Liebe rustet, so
erschlafft sie bier gleichsam, und Idsst unentschlossen ihr Geschopf in einem unent-
schiedenen, weichen, unseren Augen oft gefdlligen, aber innerlich unkrdftigen und
univirksamen Zustande.'' Here he refers to double flowers, whose stamens are
converted into petals. Thenhe goes through the different leaf-forms: the seed-
lobes are thick because they are filled with raw material, while the leaves of
the stem, and still more of the flower, become finer and finer on account of
the fact that only finer saps penetrate into them. Another idea that, besides
the saps of various degrees of tenuity, plays a conspicuous part in Goethe's
vegetable physiology is ''Anastomosis,'' by which he apparently means the
intercommunication between various parts of plants; the idea, however,
remains obscure and is certainly not made any clearer by the fact of
x8l THE HISTORY OF BIOLOGY
fertilization's being called "eine geistige /4«^j-/-o;;z(?j-e." Even florescence is caused
by "geistige Krafte," since the operation of these forces preponderates over
the raw saps that form the leaves. And, finally, Goethe develops a theory of
germination, according to which the seeds and the leaf-buds are compared.
Goethe himself admits that his metamorphosis work does not contain any
really original observations; the metamorphosis theory itself occurs in
Linnasus's Philosopbia botanica, in which, under the heading " Metamofphosis
vegetabilis," the bud, the leaf, and the flower are analysed and the leaves in
their various transformations identified. Goethe, who admits that he had
read that work, nevertheless claims to have "discovered metamorphosis";
by this he cannot reasonably mean anything else than its philosophical
side — the theory of the ideal type, according to which the leaves are trans-
formed. The "Geistige" that so frequently recurs in the treatise on metamor-
phosis is explained by the fact that it was this that Goethe considered to be
the essential, as also did Herder, of whose theory the plant-metamorphosis
doctrine is most reminiscent. For it is romantic philosophy from beginning
to end; it bears no resemblance whatever to modern natural research.
His theory of colours
It was in the course of his journey to Italy, as a result of the impression made
upon him by the southern vegetation, that Goethe first had the idea of his
metamorphosis theory. During the same journey he had also studied, in
company with some artists in Rome, the laws of colour-combination and its
effect upon the sight. Not content with the results obtained in this respect,
he resolved upon his return home to devote himself to the study of colours
from the physical point of view as well. He procured a prism and with its
aid studied a number of light and colour phenomena. These he described
with great lucidity and accuracy in a work entitled Beitrdge %ur Optik, pub-
lished in 1791. He had, however, made one or two observations — that the
centre of a large white surface viewed through a prism remains white and
that a black line on a white ground is resolved by the prism into colours —
which he considered it impossible to explain by the Newtonian laws of
optics. It is true, some physicists in the piofession who read his book ex-
plained the phenomena to him in the light of Newton's theory, but Goethe
does not appear to have been much edified by it. Then Schelling took up the
question. As mentioned above, to him light represented the "absolute
identity," and he enthusiastically hailed Goethe as a liberator from New-
ton's detestable spectral theory. The poet, who was extremely sensitive to
applause as well as to criticism, was thereby entirely won over to the new
natural philosophy and felt encouraged to go on with his optical investiga-
tions in the hope of creating a new "colour theory" in place of Newton's.
After years of preparation he finally (in 1808) published his Farbenlehre —
the greatest of his scientific works and the one that he himself valued most
SEVENTEENTH AND EIGHTEENTH CENTURIES iS}
highly. The theory of colour that he develops in this work agrees entirely
with Schelling's polarity theory. All colour-effect is derived from a "primal
phenomenon" — namely, the contrast between light and darkness; between
these two stands as a connecting link ' 'das Trube. ' ' When pure light is broken
up by a prism, it is disturbed by the action of the glass, and from this arise
the spectral colours. That these colours arise through the disturbance of the
light Goethe tries to prove, infer alia, by the fact that the sun, when viewed
through a darkened glass, appears red. Newton's view that the pure white
light actually arises as a result of the combination of the various colours in
the spectrum puts Goethe into a furious passion; he goes through Newton's
optics point by point and provides them with marginal notes which are as
irrational in content as they are scurrilous in tone. The coarsest expressions
in the vocabulary denoting stupidity and dishonesty are lavished on page
after page of the work. Goethe has here — to his own discredit — adopted
his admirer Schelling's polemical vocabulary, while his general attitude
towards Newton displays in a deplorable manner the narrow limitations of
even a universal genius. Goethe deserves no place in the history of optics.
Nevertheless, Goethe was not entirely wrong when he considered the
colour theory to be his best natural-scientific work. In fact, it contains a
section in which Goethe's finest gifts as an observer of nature are given full
play as nowhere else — namely, the chapter on "physiological colours."
In this chapter, as well as here and there in other parts of the work, are a
large number of observations of subjective colour-perceptions, recorded with
all the exactness of a scientist and with the keen insight of an artist. These
detailed observations concerning colour harmony, colour contrasts, com-
plementary colours, and other optico-physiological phenomena, attracted
great attention even among his contemporaries; they resulted in continued
research by scientists possessing professional knowledge of quite a different
type from that of their model, and even in our own day, when mental phys-
iology has become a specialized science, they have won justifiable recog-
nition. In no field has Goethe so nearly approached the spirit of exact
natural research as he has here, and that, too, in spite of the false theory
for the sake of which this subsequent research work was carried out.
During the remaining years of Goethe's life the colour theory absorbed
most of his scientific interest; in fact, in his old age he ended by valuing it
above all his poetic works. This was manifestly due to the fact that as a
poet he considered himself neglected; his neo-romantic proteges, Schelling's
friends, had come to dominate public opinion, and though they always
treated him with courtesy, they wounded his feelings by placing their own
quite mediocre leaders on a level with him. On the other hand, they loudly
praised his scientific speculations; there grew up quite a school of scientific
students of natural philosophy who looked up to Goethe as a prophet —
i84 THE HISTORY OF BIOLOGY
all these being factors conducive to continued scientific production. The
colour theory was reprinted and amplified; the anatomical writings of his
youth were published and largely added to, the ideal primal type and Schel-
ling's polarity theory reappearing in several variations. In these works he
tries to find a primal form for the anatomical details, just as he did a primal
phenomenon in optics; he invented the word "miorphology," knowledge of
form, and it still survives in modern science, although, it is true, in an en-
tirely different meaning from that which Goethe originally gave it. Of these
works may be mentioned an article on the six vertebra; composing the
cranium. Ten years before, Oken had expounded a similar theory, which will
be referred to later on; he has thus the prior right to the idea and declared,
moreover, that he mentioned it to Goethe in the course of conversation,
which, however, the latter emphatically denied. It is scarcely possible now to
find out exactly what happened; besides, it is of not very great interest
nowadays, as the whole theory is out of date.
Spiral theory
The article "tjber die Spiralfenden^' der Vegetation belongs to Goethe's last
years. Both in its idea and in its method this article is one of the most ec-
centric imaginative creations of the romantic philosophy, but for this very
reason it aroused great enthusiasm amongst the supporters of that tendency,
whilst those who hoped to see in Goethe a modern natural scientist passed
it over in complete silence. According to this article,^ the plant is composed
of two indissolubly connected "tendencies": the vertical, which represents
the eternal essence, and the spiral, which represents the nourishing, the culti-
vating, the reproductive. The latter tendency, naturally represented by the
spiral vessels, is given a number of utterly incomprehensible definitions: "das
Spiralsystem ist abschliessend, den Abschluss befordernd. Und Z}(-'ar auf gesetxjiche,
vollendete Weise. Sodann aber auch auf ungesefzliche, voreilende und vernichtende
IVeise." The aquatic plant Vallisneria, in particular, the male flower of which
grows straight, while the stalk of the female flower after fertilization con-
tracts into a spiral, is analysed in connexion therewith, the result being that
as a general rule the vertical represents the male in the plant, and the spiral
the female, which is confirmed by the ancient metaphor of the tree and the
vine-tendril which winds itself round it, as a symbol for the masculine and
the feminine in life. With this glimpse into the innermost soul of existence
Goethe concludes the "spiral" article, which was written six months before
he died, so that after all it is the poet in the old philosopher Goethe that has
the last word, which is only right, as the need for a deeper and wider poetic
view of nature was undoubtedly the true reason for his coming to grips with
the study of nature.
1 Here, too, Oken had previously dealt with the question; in his natural philosophy
there is a fantastic exposition of the spiral vessels in plants.
SEVENTEENTH AND EIGHTEENTH CENTURIES 2.85
His influence
Goethe's posthumous reputation as a natural scientist has varied generation
after generation. From the exact scientists of his own age he met with but
little encouragement — optical science in particular, which was just in his
time making brilliant progress in the hands of Fresnel, Wollaston, Brewster,
and others, naturally left him far behind — while, on the other hand, the
entire school of natural philosophy saw in him an inimitable master. And
with good reason, for as a matter of fact he did more than any other to pre-
serve the reputation of natural philosophy. When, however, this tendency
w-as finally abandoned and mercilessly given up to ridicule, Goethe was ac-
corded far more indulgent treatment; his great authority as a poet and man
of culture exempted him from harsh treatment at the hands of scientific
critics. Goethe's morphological speculation received a new lease of life
through the coming of Haeckel; for reasons that will be explained later on,
Haeckel expressed a boundless admiration for Goethe and regarded him as
one of the foremost precursors of Darwinism. On his authority both the
general public and literary history have since willingly accorded Goethe,
who in other respects has left so many marks on the cultural development
of our time, the further honour of being a natural scientist in the modern
sense of the term. Nevertheless, his biological writings have certainly been
more admired at a distance than read in the original, a fact that has no doubt
contributed in the long run towards concealing their true quality.^ Goethe
was no exact scientist, but a romantic natural philosopher; in that capacity,
however, he has also exercised an influence, which must not be underesti-
mated; his psycho-physiological observations and speculations have formed
the basis on which men like Johannes Miiller and Purkinje have built up
their work, and although Goethe may have had no eye for comparative
morphology in the modern sense, yet many an eminent anatomist of later
date has been induced by Goethe's ideas to devote himself to a comparative
study of form that has proved of benefit to science. Goethe takes his place in
the history of biology as a stimulating force; his influence was, it is true,
both good and bad, but by no means inconsiderable.
- So far as is known, no separate edition of Goethe's scientific writings has existed until
recent times; those desirous of studying them have had to have recourse to the somewhat ex-
pensive editions of his Sdmtlkhe IVerkt. An edition of these writings was published some years
ago by R. Steiner, who, as is well known, made Goethe's conception of nature the basis of
his " anthroposophical " theory of existence, which does not in the least accord with modern
natural science.
CHAPTER XIV
NATURAL-PHILOSOPHICAL BIOLOGY
I. Germany and Scandinavia
Character of the natural philosophy of the time
THE DIRECTION taken by natural-philosophical thought that has been
described in the foregoing has played a very important part in the
cultural development of the world and not least in the science of
biology. There were, of course, during the natural-philosophical period a
large number of scientists who were not at all, or only very slightly, affected
by the speculative tendencies of natural philosophy, while others, it is true,
embraced its tenets either temporarily or permanently, but at the same time
carried out research work in exact natural science with lasting results. The
work of these scientists will be recorded later on; in the present chapter we
shall devote our attention to a group of scientists who applied themselves
entirely to a speculative explanation of nature and sought to incorporate in
it all the known facts about nature that they considered necessary and attain-
able, or who, at any rate, in a more or less pronounced way gave themselves
out as champions of such views. It was in Germany and Scandinavia in
particular that these faithful disciples of Schelling and the other idealist
philosophers won for a time extraordinary success and managed to present
to the public, and particularly to the universities, their master's theory, with
amendments of their own, as the only true natural science. The causes of this
phenomenon, which must seem strange to us, as it was to earlier generations,
were manifold. The universal cultural tendencies that favoured romanticism
in general — disappointment at the failure of the efforts to win liberty under
the revolution and weariness after the great wars of independence — natu-
rally also played an important part in the development now under discussion;
again, the interest in mysticism that spread far and wide at the close of the
eighteenth century and was cultivated by the numerous brotherhoods, un-
doubtedly had a great influence. The possession of some form of knowledge
that is unattainable for the majority has always been an attractive prospect
for human egotism — now it was possible for the professor of philosophy or
natural science at the university to present to his hearers a theory which at
any rate had the advantage of being incomprehensible to the uninitiated;
i86
o
N
%
Pi
o
u
<
<
1-1
SEVENTEENTH AND EIGHTEENTH CENTURIES iSy
when, moreover, according to its tenets, the elect possessed the right to be
hailed as geniuses, it is easy to realize the enthusiasm which the new wis-
dom evoked. The result was that there developed at the universities such
student insolence as had never existed since the days of scholasticism; and it
survived for a long time, especially at the academies situated in the provincial
towns, where neither masters nor pupils had much contact with practical
life. This very academic isolation explains to a certain extent how it was
that these theories, so out of accord with reality, could survive for so long,
and explains the fact that they were localized in Germany and Scandinavia,
while in western Europe, with its more lively and practical activities,
speculation at least adopted more dispassionate forms.
One of the most notable and influential personalities in German natural
philosophy was Lorenz Oken (1779-185 i). He was of south German peasant
stock, his family name being really Ockenfuss, and was brought up in poverty,
though he managed to obtain a school education and afterwards studied
medicine, eventually becoming a doctor, in 1804. Medicine, however, was
not of very great interest to him; at an early age he had formulated a natural
philosophy of his own. After having maintained himself under severe pri-
vations at several universities, he was in 1807 appointed assistant professor
at Jena, where he published as his inaugural address his paper on the subject
of the cranium's being composed of several vertebra;. This resulted in his
falling into disfavour with Goethe, which caused him considerable un-
pleasantness, all the more so as he was of a passionate nature and found it
difficult to exercise discretion in his behaviour. Being an ardent German
patriot, moreover, he was enthusiastic for his country's unity and was con-
sequently suspected by the authorities in the reaction after the War of In-
dependence, for which he zealously agitated. At last, in 1819, he was forced
to resign, although he had the support of the entire University; he was for a
time without an appointment, but afterwards became a professor at Munich;
there too, however, he was unable to get on with the authorities, so that he
was glad to accept a post in Zurich in 1831. He carried on his work there,
respected and esteemed, for the rest of his life.
Oken's activities were many-sided and his influence upon the develop-
ment of culture considerable. For many years he published the journal
his — the name is characteristic of his half-mystical philosophy — which
became a focus for the scientific life of Germany; with great impartiality it
accepted papers by scientists of different camps; the discussion of problems
was encouraged, and prizes offered for solutions, with the object of promoting
scientific research. Oken took the initiative in another idea which has proved
of value to the future of science; he organized meetings of scientists for the
purpose of exchanging views and encouraging sociability. Thus it was he
who originated those gatherings that are so much appreciated in our own
l88 THE HISTORY OF BIOLOGY
day and which he himself stimulated and developed by his lively personality
and his keen interest in the whole domain of thought. Finally, by his writings
he promoted also an interest in the study of nature; his Allgemeine Naturge-
schichte fur alle Stdnde is a compilation of a very high standard of excellence,
based on comprehensive material, which has widely increased the knowledge
of and interest in the study of nature.
Oken's natural philosophy
Oken's own contributions to exact natural science are, on the other hand, of
but little importance. In his youth, before going to Jena, he carried out an
investigation into the development of the intestine in the embryo, which
contains a number of sound observations, though his conclusions were partly
drawn from principles that were not very successfully thought out. Oken, in
fact, considered himself above all a natural philosopher and set great store
by his best work, Lehrbuch der Naturphilosophie, which he rewrote twice. He
was, however, not very learned as a philosopher; his conclusions were as
fantastic as Schelling's, but had not even the latter's small degree of formal
consistency and logic. Nor did he possess Goethe's poetical imagination,
and his speculations were therefore as grotesque as they were irrational. In
particular, the first part of his work, which, strikingly enough, is called
"Theosophie," is simply extraordinary; its first sentence runs: "Die hdchste
mathematische Idee oder das Grundprincip aller M.athematik ist das Zero = 0."
Then we learn that God and the world = O H , while God alone or
the primal idea = O, and space is O = + O -. When we come to the living
creatures, the whole is certainly somewhat closer to facts; organic life is
derived from a primal slime, which is described as "oxydierter, gewdsserter
Kohlensfojf," and which had its origin in the sea, whence all life comes. Life
is formed of three "entelechia;" : magnetism, chemism, and respiration. In
regard to plants, Oken, like Goethe later on, speculates upon the spiral
ducts; to Oken they are ' 'das Lichtsystem in der PJJanze. ' ' The parts of the plant
correspond to the four elements, the root being the earth-organ, the stem
the water-organ, the leaf the air-organ, and the flower the fire-organ. With
regard to the animal kingdom, all animal life is derived from a follicle; there
are four consecutive formations thereof — the point-, the ball-, the fibre-,
and the cell-formation. The organs of animals constitute special systems,
first "pfianzliche" (namely, intestine, gills, veins); then "thierige" (bony,
muscular, and nervous systems). Moreover, the animal is composed of
"Hirnfier" and "Geschlechtstier," both of which possess organs that corre-
spond to one another, as, for instance, lung and bladder, mouth and rectum,
thorax and pelvis. The animal kingdom in its entirety is regarded as one
mighty animal, the various parts of which correspond to different animal
forms; the lowest animals have only intestine, as the polypi; then come such
as have intestine and skin — snails and insects; finally, those having in-
SEVENTEENTH AND EIGHTEENTH CENTURIES 189
testines, skin, and flesh — the mammals. Further quotations vvould be super
fluous. We find in the above a great deal of ancient mysticism, such as the
mysticism of numbers — recurring groups of three and four — the comparing
of the animal kingdom to a great body, reminiscent of Swedenborg's specula-
tions, and, finally, the strange idea of figures at the beginning of his work. At
the same time we find here some occasional idea that recalls biological theo-
ries of our own day, as, for instance, the follicle-shaped primal animal, the
idea of the sea as the origin of animal life. Oken was undoubtedly a man of
ideas, many of which might be of value to the future; his unbridled imagi-
nation, however, made him a warning to the succeeding generation and an ill-
directed example of what the results of natural -philosophical speculation
may be.
Christian Gottfried Daniel Nees von Esenbeck (1776-185 8) forms a
parallel to Oken in the sphere of botany. From the south of Germany, like
Oken, he was the son of a public official; he studied medicine at Jena, where
he was won over to Schelling's philosophy and came into contact with
Goethe. Having completed his studies, he settled down on an estate that he
had inherited and there worked as a private scholar until the year 1818, when
he was appointed professor of botany at the newly-founded University of
Bonn. He established the botanical institute and gardens there and wrote a
number of books on both botany and natural philosophy. Later, having
been appointed professor at Breslau, he began by doing some successful work.
In old age, however, the romantic natural philosopher became ultra-radical;
he took part in the labour movement, zealously supported ideas for the re-
form of Christianity, and worked hard in theory and practice for free mar-
riage without State co-operation; the end of which was chicanery, dismissal,
and death in poverty. The Labour Union at Breslau, whose chairman he was,
followed him to the grave.
As a classifier of plants, Nees von Esenbeck has acquired a distinguished
name; he has won special fame for his tropical floras, dealing with the phan-
erogamous plants of the Cape and of Brazil; his works on the cryptogams,
on lichens and hepaticas, alga; and fungi were also at one time highly thought
of. He himself, however, set greater store by his natural-philosophical spec-
ulations. In his Lehrbuch der Bofanik, which he dedicated to Goethe, he carried
the latter's metamorphosis theory to the uttermost extreme. To him the
leaf is a kind of symbol for the plant as a whole; the entire vegetable world
is to him one mighty leaf, just as the animal world was to Oken one mighty
animal. Even in the vegetable world the number three plays an important
and mystical role and is the basis for a good deal of play upon words. Polarity
in the style of Schelling recurs once more; the fungi represent the north and
the plants the south, the animals midnight and man noon, while the chemi-
cal components of plants are dealt with just as arbitrarily. The colours in
Z90 THE HISTORY OF BIOLOGY
the vegetable kingdom are of course treated in accordance with Goethe's
theory of colour. Even the spiral vessels give rise to a number of speculations,
although Goethe's theory had not yet been published. The spiral theory was,
in fact, developed later by a large number of botanists who, like Nees von
Esenbeck, could be quite rational collectors and systematists, but who at
the same time gave play to a wild and reckless imagination on the subjects
of the spiral and polarity, till at last, in this sphere also, exact research
claimed its due.
One of the last survivors of Germany's natural philosophers is worthy
of mention — Carl Gustav Carus (1789-1869). Born in Leipzig, he became
professor of comparative anatomy there in 181 1 and afterwards professor of
gynaecology and court physician at Dresden. As a doctor he was an eminent
specialist and, besides, a man of unusually varied interests; a personal friend
of Goethe's, he had a truly artistic nature and was himself a good painter
and writer on art, as well as a comparative anatomist and natural philoso-
pher. He experimented in comparative osteology, insect anatomy, and zo-
otomy. His comparative anatomy (of 182.8) stands to a certain extent on the
border-line between the contemporary and a more modern conception of that
science. Carus, for instance, no longer goes in for the plus and minus signs
with which his predecessors wasted their time without in the least solving
their problems; to him, indeed, nature is an expression of an idea, and life is a
flux, and the three-grouping recurs here and there; but he is at any rate able
to describe an organ or a system of organs without at once becoming in-
volved in sheer incomprehensibilities. He sets up an animal system arranged
in circles, one inside the other, with the protozoa outermost and man inner-
most, and with not very successful descriptions, but, on the other hand, he
gives a comparative account of the nervous system throughout the animal
kingdom that is arranged clearly and in an exact form throughout. In 1861, as
an old man, Carus summarized his ideas in a work entitled Nafur und Idee,
which certainly strikes a curious note, considering the advances that natural
science had already made by that time. Here we find ether regarded as the
primal substance and the essence of all chemical elements; after which we are
told that ''die Urbandlung des Athers ist Leben." The nervous system is the
central force in the animal kingdom, like the primal fire and the electrical
principle in the earth; the universe is an infinite sphere whose centre point is
everywhere and whose periphery is purely ideal. Again, animals arise out of
a sphere, the ovum, and develop through new spheres* being added to the
original; the senses are drawn in the corners of a pentagram inscribed in a
circle. When this work was published, Darwin's theory of selection had been
known for two years. Thus the last representative of natural philosophy in its
most extreme form survived up to modern times.
The neo-romanticist natural philosophy was brought to Scandinavia
SEVENTEENTH AND EIGHTEENTH CENTURIES 191
by Henrik Steffens, who, although he has played little part in the develop-
ment of biology, nevertheless deserves mention here owing to his importance
in cultural history. Born at Stavanger, in Norway, he studied in Copen-
hagen and Kiel and after a period of wandering came to Jena, where he be-
came an enthusiastic admirer of Schelling, a friend of Oken, and a natural
philosopher heart and soul. He returned to Copenhagen in 1802. and for a year
or two lectured at the University, but as he obtained no permanent post
there, he accepted an appointment in Germany, where he remained for the
rest of his life. During the years he spent in Denmark he exercised great in-
fluence by his enthusiastic promulgation of natural philosophy, although his
exaggerations aroused doubts in the minds of the Danes, which were not
perceived by his less critical friends in Germany. His principal work on
natural philosophy dealt with the internal natural history of the earth; in it
he seeks to prove, inter alia, that the various strata of the earth are sections
of a galvanic element. Of importance to the history of biology was his theory
of the origin of the circular coral islands; he believed that they grew up on
the edge of volcanic craters in the ocean, and this theory was accepted as true
by many, until Darwin disproved it by his well-known investigations into
the subject.
In Sweden natural philosophy was embraced by the famous Carl Adolf
Agardh (1785-1859), known as one of Sweden's most many-sided geniuses.
He was a native of Scania, matriculated at Lund, and eventually became
lecturer in mathematics and professor of botany and economics at that
University. He ended by being Bishop of Karlstad, after having won renown
as a botanist, mathematician, national economist, priest, and politician.
Only his sphere of activity as the first belongs to this narrative. At Lund
Agardh became acquainted with the Linn^an system of plant classification,
and in the course of journeys in Germany he came to know Schelling and
natural philosophy. His most lasting fame he has won as one of the founders
of alga; classification; much of the system that he created still exists today.
He has also made valuable contributions in connexion with plant classifica-
tion as a whole; in particular he was one of the first to pronounce against the
sharp difference that had hitherto been held to exist between phanerogams
and cryptogams. His general views with regard to animate nature he has col-
lected in a handbook of botany published in the years 1818-32., the first part
of which he dedicated to Schelling. This first part, entitled "Organography,"
contains also traces of the influence of natural philosophy; nevertheless
Agardh displays a degree of caution in speculation that is in favourable
contrast to the rashness of his German master. Thus he at once declares on
the first page that natural objects cannot be exactly defined on a logical basis;
he believes that we have to content ourselves with establishing in each what
is the most common phenomenon or the most usual form, without venturing
X9X THE HISTORY OF BIOLOGY
to lay down rules having no exceptions. And when he quotes Nees von
Esenbeck's above-mentioned comparison between natural objects and the
points of the compass, he does so with the — certainly mild — reservation
that "this method of philosophizing is beautiful but obscure." On the other
hand, he can hit upon such eccentric ideas as that the human hands may
represent leaves between which the head sits like a bud; and the metamor-
phosis theory likewise leads to a number of extremely bold comparisons
between the stages of development in plants and animals. But there is observ-
able throughout a keenness of observation in points of detail which proves
that the teachings and example of Linnsus had not lost their influence in
his own country. The second section, on vegetable biology, treats of the
manifestations of life in plants and is on the whole more exact than the
former section.
Another important representative of natural philosophy in Scandinavia
was Israel Hwasser (1790-1860). He was the son of a priest at Alvkarleby,
and after being educated at home he matriculated at Upsala, where in 1812.
he took the degree of doctor of medicine. Five years later he became pro-
fessor of medicine at the academy of Abo, where he exercised considerable
influence. The medical education there, which had fallen into decline, was
improved by him in regard to both the number of students and the standard
of the knowledge imparted, while he himself succeeded in gathering about
him friends and pupils who took part in his idealistic labours. In 1830 he
applied for and obtained a professorship in Upsala, where he afterwards
worked until his death. The reason for his transfer was that he wished to
counteract the scheme just then being proposed for removing the medical
school to the Carolinian Institute in Stockholm, which, he argued, was at
variance with the principle of the connexion of ideas in scientific education.
But he maintained his interest in Finland throughout his life. He was in
everything an ideally minded personality who in speech and writing as well
as in private company was a zealous supporter of patriotism, loyalty, and
clean morals and in this respect exerted great influence on the young people
in the University. His scientific activities he desired also to place entirely
at the service of moral ideals. He was, it is true, a natural philosopher, but
he did not approve Schelling's speculations, especially the attempt to con-
struct nature out of an idea; on this attempt he passes the weighty, and on
the whole correct, judgment: "It was a scientific extravagance pushed to
extremes, which in the minds of some of those who took part in it seemed
to have been fostered and supported by a pride nearly akin to madness."
On the other hand, he was a great admirer of Sydenham and still more so
of the French anatomist Bichat, who will be described later on. His whole
conception of life in nature is characterized by his deeply ethical aims. He
is dissatisfied with those who, starting from the lower, would try to under-
SEVENTEENTH AND EIGHTEENTH CENTURIES 193
Stand the higher, whether they do so by means of theoretical construction
or practical investigation. In general he despises any dealings with matter
as such and denies its indestructibility. To him life is a magnificent process
of ethical refining; to his mind the development of the individual represents
the selfish element, while reproduction, which entails self-sacrifice for the
welfare of the race, is the highest element in the organism. He is zealous
also for an ideal conception of love and marriage and for that reason dis-
approves of Goethe's sensual love-poetry and therewith of all that Goethe
did. His theory of disease as self-destruction in the individual is in accord-
ance with this, his conception of life. Nevertheless, he by no means dis-
approved of practical medical education, though he himself had little to
do with it, owing to bodily clumsiness, and, as he himself declared, conse-
quent laziness. As will have been seen from the above, his biological theory
was no less inconsistent with true nature than Schelling's, but at any rate
it had the advantage of assuming as its chief mission the improvement of
morals, which Schelling's certainly did not. It was in any case Hwasser's
personality that exercised the best influence; it is mostly for this that he is
remembered today.
In Finland Hwasser's natural philosophy was maintained after his death
by his faithful friend and pupil Immanuel Ilmoni (1797-185 6), who shared
his ideas and warmly defended them. With the passing of these two men
natural philosophy disappears from the universities of the North, where, as
a matter of fact, it had never made the same progress that it had done in Ger-
many; biology was mostly carried on throughout the natural-philosophical
period on Linnasn principles; moreover, men like Berzelius and Anders
Retzius were working for exact natural research, and natural philosophy
had no rivals in Scandinavia to compete with them.
2.. England and France
Cultural development in ivestern Europe
Natural philosophy has by no means played the same part in the two lands
of culture in western Europe as it did in Germany. The reason for this may
be sought ultimately in the national character of their peoples, Englishmen
and Frenchmen always showing themselves less speculative and more in-
clined to direct their energies towards practical aims than the Germans.
And indeed practical functions were far more attainable in those uniformly
governed and well-organized western-European kingdoms than in the di-
vided and politically disillusioned country of Germany. The reaction against
the opinions of the eighteenth century sought and found its expression, both in
England and France, in politics, both of Church and State, and in literature.
194 THE HISTORY OF BIOLOGY
while science was abk to continue its work undisturbed by any serious
revaluations of the old standards. Speculative tendencies had indeed existed
in these countries at an earlier period — Buffon is, of course, the most bril-
liant example — and the theories of the true natural-philosophical age often
appear as continuations of those tendencies, while, on the other hand, they
serve as the transition between the past and exact science in the nineteenth
century. Nevertheless, even during this period there arose scientists whose
speculations had more in common with the natural philosophy so far de-
scribed, and who to a certain extent actually had direct points of contact
with it. It is now proposed to give one or two examples of speculations of
this kind, while such scientists as seem to stand in a more direct relation
to modern biology will be discussed at the beginning of the next section of
this work.
In England natural-philosophical speculation has been a familiar prac-
tice from early times. Many of its pioneers have combined a wealth of origi-
nal ideas and theories for explaining natural phenomena with somewhat
unsystematic methods of thought and experiment. More or less gifted authors
of this type there were in abundance, especially in the numerous circles of
private scholars in England. In this category may be included Erasmus Dar-
win (i73i-i8ox), whose speculations caused a sensation in his day, not only
in England, but also on the Continent. Born at Nottingham of an old stock,
he devoted himself to medicine, studied in Cambridge and Edinburgh, and
finally practised as a doctor at Lichfield. He is described as very original,
vigorous and somewhat coarse-grained, honest and straightforward; besides,
he was a keen worker and well thought of in his profession, kind towards
the poor, and an ardent supporter of temperance. He had many children;
one of his sons was the father of Charles Darwin, and a daughter became
the mother of Francis Galton, the student of heredity. Apart from his own
profession, Erasmus Darwin was an indefatigable author and wrote a great
number of papers for the Royal Society and also published one or two col-
lections of poems, with which he himself was highly satisfied, but which
fell a victim to the ridicule of his contemporaries and the neglect of posterity.
The work, however, which alone has made his name memorable is his
Zoonomia, an attempt to find out the laws of organic life, which was pub-
lished in 1794 and was translated into several European languages. ' It
excited a good deal of attention at the time; the German natural philoso-
phers in particular have quoted it with appreciation. In modern times, how-
ever, it would certainly have caused but little notice had not the author
been the grandfather of Charles Darwin. At first sight it gives the impression
of being a most extraordinary conglomeration of diverse notes on scientific
^ The author has not succeeded in coming across the original of this work, but has been
compelled to have recourse to Brandis's German translation.
SEVENTEENTH AND EIGHTEENTH CENTURIES 2.95
and medical subjects; but on closer inspection we find that it deals with a
number of problems that engaged the minds of the author's contemporaries,
although from a curious point of view.
Erasmus Darwin's natural philosophy
The work begins with the assertion that spirit and matter are the foundations
of nature; then life is defined as being due to a special force, which, after
Haller, is called irritability and by aid of which all life-phenomena are ex-
plained in a somewhat peculiar manner. All manifestations of life, both phys-
ical and psychical, are due to the contraction of fibres, which is induced by
irritation; by "idea" the author understands a contraction of the fibres that
form the direct sensory organs. La Mettrie himself could hardly have ex-
pressed himself more materialistically, but Erasmus Darwin is by no means
a materialist as the term was understood by his own age; true, he refers to
the sceptic Hume's inquiries into cause and effect, but at the same time cer-
tifies his invincible faith in the Bible, quoting verses of the Psalms on the
wisdom of the Creator and citing the words of Moses in the Book of Genesis
touching the creation of Eve from Adam's rib as a proof of his own theory
of reproduction. This latter theory, more than anything else that Erasmus
Darwin wrote, attracted great attention, which indeed it undoubtedly de-
served. To a certain extent it is reminiscent of Caspar Friedrich Wolff's theory
of evolution — that is to say, in so far as it is markedly epigenetic. Whether
the author had recourse to Wolff is not clear from his work — possibly he
had on second-hand information. But, above all, his theory is pronouncedly
animalculistic; just as the whole basis of his theory of life rests on the as-
sumption of "irritable" fibres being the basic substance of all living things,
so the origin of the embryo is a " filament, ' ' which is derived from the father
and to which the mother only gives nourishment. As a result of this latter
the embryo grows, by no means, however, as a result of the development
of ready-formed rudiments, but through the addition of fresh matter. The
author seeks to disprove the preformation theory by arguments both serious
and facetious; Bonnet's incapsulation theory in particular seems to him ex-
ceedingly ludicrous; the dimensions of the infinite number of embryos con-
tained one inside the other remind him of how St. Anthony was tempted
by twenty thousand devils, all dancing on a pin-point. In further confirma-
tion of his epigenesis theory he declares that the male "filament" which
gives rise to the new individual is manifestly influenced by the nourishment
which the mother provides; any resemblance between mother and child is
due thereto, as is particularly shown in bastards. And still further: the con-
ditions under which the parents live clearly influence the character of their
offspring, as is proved by the new varieties obtained from domestic animals;
organs which the animals need are produced by irritation in the parts of the
body which form them, and they are afterwards inherited by their progeny.
196 THE HISTORY OF BIOLOGY
Thus stags have got horns, and cocks spurs, while fighting for their
mates. In fact, one is justified in assuming that all living creatures, different
from one another as they now are, nevertheless originate from one and the
same "primal filament," whose offspring have become changed as a result
of different conditions of life, this being confirmed by the existence of tran-
sition between all animal and vegetable forms, both higher and lower. This
theory undoubtedly sounds to a certain extent "Darwinian," but what dif-
ferentiates the grandfather's doctrine from the grandson's is the problem
each set out to solve: Erasmus Darwin really had no interest in the origin
of species; with him it was a question of obtaining as strong evidence as
possible for the epigenesis theory, which was a marked feature of his specu-
lations. Nor indeed was it this side of his work that evoked contemporary
interest; it was rather his speculations on the subject of the life-force and,
further, his theory of irritability and his observations of sense-impressions,
which he made with a view to confirming the latter theory and which in
a certain degree foreshadow Goethe's. All this afforded special interest to
the German natural philosophers, who not infrequently refer to his writings.
He was entirely forgotten by the succeeding generation; in fact, it was not
until after Charles Darwin had become world-famous that interest in Eras-
mus revived, when an attempt was made to see resemblances between his
speculations and those of his grandson. Some resemblance there certainly is,
but it is undeniable that the originator of the theory of selection had worked
with entirely different qualifications from his grandfather's in order to pro-
duce a universal theory of the evolution of life.
In France biological speculation during the latter half of the eighteenth
century was essentially governed by Buffon's ideas. He and his friend Dau-
benton had, it will be remembered, carried out comparative investigations
into the anatomy of various animals, especially the bone-structure of mam-
mals. During the following epoch also French scientists were keenly occupied
in investigations of this kind, and natural philosophy, to which such investi-
gations were at that time referred, thereby assumed a more defined and prac-
tical character than in Germany. Moreover, its exponents stand out far
more clearly as the precursors of modern biology and deserve to be discussed
in connexion therewith. In particular, he who is regarded as the foremost
natural philosopher that France has produced — Lamarck — seems, in view
of the great influence he has exercised on modern research, to be most worthy
of record amongst its pioneers. A natural philosopher who, on the other
hand, is far more closely associated with the German speculation, and who
was actually connected with it, was Geoffroy Saint-Hilaire. He may there-
fore suitably be described in this context.
Etienne Geoffroy Saint-Hilaire was born at Etampes, near Paris, in
the year 177^, the son of a public official. His father had him educated for
SEVENTEENTH AND EIGHTEENTH CENTURIES 197
the priesthood and actually procured him a benefice, but at the same time
allowed him to follow his bent for natural studies. Highly gifted, impulsive,
and passionate, young Geoffrey became deeply engrossed in the study of
chemistry, crystallography, and anatomy. During the Revolution he dis-
tinguished himself by rescuing a number of priests from death during the
massacres in September 1791 at the risk of his own life. In spite of this he
was appointed in the following year by the Revolutionary Government pro-
fessor of zoology at a newly-founded educational establishment and at once
became known for his brilliant energies and success. Cuvier, who was at
that time still quite unknown, was promoted to another professorship under
his recommendation. Eventually Cuvier was to rise above the head of his
patron, but for the time they collaborated with success in the sphere of com-
parative anatomy. When Bonaparte made his famous expedition to Egypt,
Geoffroy accompanied him as zoologist and succeeded in making there a
number of splendid collections, which he later, thanks to his resolute action,
prevented from falling into the hands of the English. The result was that,
upon returning to Paris, he won still further honours. He won less glory
in an expedition that he m.ade to Portugal, in whose museums he brought
together ' ' collections ' ' at the command of Napoleon on behalf of the French
State. His later years are mainly characterized by an increasing rivalry and
enmity with Cuvier; they were very largely contrasts to one another. In his
old age he became blind and finally also paralytic. He died in 1844.
Geoffroy' s comparative anatomy
The comparative anatomy introduced by Buffon and Daubenton was enthusi-
astically embraced by Geoffroy and Cuvier. The development to which this
science attained, chiefly thanks to Cuvier, who made it one of the most
important foundations of modern biology, will be described in the following
section. Nor, indeed, was GeofFroy's contribution towards the progress of
this science without its significance, but at quite an early age there developed
in him a fancy for imaginative speculation, which justifies his being placed
in the category of theorizing natural philosophers. It has been thought pos-
sible to trace the influence of German natural philosophy, especially Schel-
ling's,- in this tendency of his, but as it was not until later that Schelling's
writings were translated into French, one can hardly suppose that there was
any direct influence from that quarter. These fantastic speculations were cer-
tainly characteristic of the age; he had predecessors in this respect even in
French literature; one need only recall the name of Bonnet. The main idea
in Geoffroy's philosophy is the existence of a common fundamental type,
beginning with the organization of all the vertebrate animals, and then for
the entire animal kingdom in general. He worked principally at the anatomy,
^ According to Kohlbrugge in his essay on Goeche; see the Bibliography at the end of
the book.
X^S THE HISTORY OF BIOLOGY
particularly the bone-structure, of the vertebrates, and here he has really
worked out a number of ideas which foreshadow results that have been
gained by modern comparative anatomy. He thus derives the auditory bone
of mammals from the cranial bone in fishes — doubtless from the opercula
and not from the bones now regarded as the starting-point. He derives the
cartilage of the larynx from the fishes' branchial arches — as has indeed
been, at least partially, done in present-day comparative anatomy. But here
at once his unbridled imagination manifests itself in the utter lack of detailed
criticism and ability to limit his speculative field; thus, for instance, he finds
a sternum in fishes, and he derives the annular cartilages of the trachea from
the gill-arches, just as he delights in making direct comparisons between
fishes and mammals in general. (In passing, he makes the truly natural-
philosophical assertion that the auditory apparatus of birds is the finest
there is, which is proved by the fact that they are so musical.) He indulges
in his wildest flights of fancy, however, when he compares vertebrates and
invertebrates. To his mind, insects and crustaceans are composed of verte-
bras, in which both apophyses and ribs can be distinguished — the joints
are vertebrae, and the extremities ribs. The shells of the tortoise and the snail
are compared, and the ink-fish is a vertebrate animal with a duplication of
the back. One realizes that this application of the theory of a common fun-
damental type for the animal kingdom must have impressed Goethe. In the
dispute that eventually broke out between Geoffroy and Cuvier — which
will be described later — Goethe loyally supported Geoffroy. When Cuvier
died, Geoffroy was left free to promulgate his own theories, which were
also adopted, in part at least, by his son, Isidore, who was likewise an emi-
nent biologist. The more critical comparative anatomy, besides eradicating
the worst exaggerations, eventually acknowledged the wealth of ideas and
the, in many respects, productive thoughts that were to be found in Geof-
froy Saint-Hilaire.
The natural-philosophical school of thought which we have endeav-
oured to describe above has had a deep influence on the development of bi-
ology. Its extravagances cannot, of course, be regarded as other than features
tending to retard the sound progress of science; time has also helped to put
them out of mind, or at worst they have been recalled only for the purpose
of ridiculing the weaknesses of an older generation. The service it has ren-
dered to humanity lies in the lively interest for the study of nature which
it evoked in the scientists of its era — an interest in striving to find law-
bound phenomena in existence. Otherwise its age certainly specialized in
speculation upon abstract ideas, as Hegel and his school would have it; but
the fact that during this period the study of nature did not disappear alto-
gether nor degenerate into a mere handicraft is at any rate due in no small
measure to natural philosophy. Many of its ideas, indeed, recur, in a more
or less revised form, in the biology of the nineteenth century, an account
of which will be given in the next section of this work.
PART THREE
MODERN BIOLOGY
BIOLOGY DURING THE FIRST HALF
OF THE NINETEENTH CENTURY
CHAPTER I
FROM NATURAL PHILOSOPHY TO MODERN BIOLOGY
I. The Predecessors of Comparative Anatomy
IN THE HISTORY of biology the nineteenth century will undoubtedly be al-
ways regarded as one of the most important epochs. With regard to the
value of the discoveries that were then made, that period can certainly
vie with the most brilliant of the periods that preceded it, and if we con-
sider the reputation which biology enjoyed in the world of culture of the
time, it was an unrivalled epoch. It is primarily the latter half of this cen-
tury, after the appearance of Darwin, that witnessed the greatest advance
that biology has ever been able to record, particularly in regard to the volume
of its discoveries. But this advance was being prepared for during the im-
mediately preceding decades in the splendid development that took place
in most branches of research — a development which in its turn was evolved
during the preceding eras out of events that have previously been recorded
in this work. Thus, in the biological science of the nineteenth century we
find elements derived from the exact scientific research that during the two
preceding centuries had sought for a mechanical explanation of the phenom-
ena in animate nature, but there were also features from the speculative
natural philosophy that endeavoured, by means of purely theoretical systems
of thought, to solve those problems of existence which exact scientific re-
search had found itself compelled to leave unexplained. As has already been
described in the last few chapters of the previous section, this natural phi-
losophy chiefly flourished in Germany and Scandinavia, but also existed in
England and France; it exercised a decisive influence upon the cultural de-
velopment of that period, and, as we shall see later on, far beyond it. During
the age when natural philosophy flourished, however, exact scientific re-
search was by no means dead; it only worked the better in peace, and, more-
over, there were of course scientists who, though convinced supporters of
301
30L THE HISTORY OF BIOLOGY
natural philosophy, nevertheless carried out exact investigations on special
subjects. A fev^^ examples of this exact natural research during the age of
natural philosophy deserve to be quoted here as representing a transition to
modern biology, which developed during the succeeding epoch.
In France, as we have already found, comparative anatomy had been
considerably developed through the work of BufFon and Daubenton. But de-
scriptive anatomy had also had a brilliant representative during that same
century in the Dane Jacob Benignus Winslow, a relative of Steno, who, like
the latter, became a Catholic and was fully naturalized abroad, but who, in
contrast to his predecessor, enjoyed an unusually long life of activity. He
died in 1760 at the age of ninety-one. His description of the human anatomy
was especially complete, particularly as regards the topographical section, and
he made the medical faculty in Paris, where he was professor, an important
centre of anatomical study. He and his immediate pupils were, however,
outshone by a man who was able to develop even the human anatomy along
comparative lines on the model of Daubenton and Camper.
Felix Vicq d'Azyr was born in 1748 at Valogne, in Normandy. He was
the son of a physician, and after being educated at school, he chose his
father's career and studied in Paris with such success that only eight years
after entering the profession he was able to give lectures there. He had no
academical career, however. He was passed over when a vacancy was filled
in the professorship of anatomy at the Jardin des Plantes in 1774, and again
upon the appointment of a successor to Buffon. Instead, he was sent by the
Board to study and stamp out serious epidemics in certain provincial parts
of France, and he wrote some valuable accounts of them. In the field of
veterinary science also he made some important contributions. Besides this
work, he held private courses in anatomy, which were very popular, and
he also collaborated in the founding of the Royal Society of Medicine in
Paris, of which he became permanent secretary; in that capacity he composed
a number of brilliant epitaphs upon past distinguished physicians, on ac-
count of which he was appointed Buffon's successor in the French Academy.
At last he was made personal physician to the King, but when the Revolu-
tion broke out, shortly afterwards, this post of honour was a cause of much
trouble and danger. His health, which had already given way under stress
of work, now broke down; finally, as a result of attending the famous feast
of the Supreme Being, under compulsion, he contracted a chill and died a
few days later.
Vicq d'Azyr's career was thus a short one, and, moreover, his energies
were divided as a result of the practical work he had to carry out in order
to make a living. On these practical activities, indeed, he expended much
labour, and his works on epidemics, veterinary surgery, and organizational
problems in practical medicine are fairly numerous. But in spite of all this
MODERNBIOLOGY 303
he found time for carrying out serious theoretical research in the sphere of
anatomy and physiology, and though the results to a large extent exist only
in the form of brief accounts published in academical proceedings, neverthe-
less he has thereby contributed largely to the development of biology. Of
one important work that he started on anatomy he managed to publish
only the first part, wherein he lays down the principles on which he con-
siders that the study of anatomy should be pursued.
Vkq d'Axyr's classification of the functions of the organism
For this purpose he takes as his starting-point, firstly comparative anatomy,
as created by Daubenton, and secondly Haller's physiological theories and
experiments. He begins by discussing the ancient division of natural objects
into three kingdoms and finds that the essential difference lies between ani-
mate beings and inanimate things; plants and animals possess common prop-
erties that stones and minerals lack. In connexion therewith he strongly
rejects the old comparison — which is sometimes repeated even in modern
times — between the growth of the organism and that of the crystal; he
points out the mathematically regular shape and homogeneous structure of
the crystal as contrasted with the rounded forms and variously constituted
systems of organisms, but above all he emphasizes the organisms' definitely
characterized functions as a peculiarity of life. These functions he divides
into the following categories: (i) digestion, (x) nutrition, (3) circulation,
(4) respiration, (5) secretion, (6) ossification, (7) generation, (8) irritability,
(9) sensibility. The existence of the various functions, together with their
respective organs, is then examined in the different life-forms; regarding di-
gestion, it is stated that man, the quadrupeds, whales, birds, and crustaceans
possess one or more stomachal cavities clearly distinct from the oesophagus
and the intestine; oviparous quadrupeds, snakes, selachians, and osseans have
a stomach in the form of a single extension; insects, worms, and zoophytes
have only one intestinal tube, and plants no digestive canal — the classifi-
cation is noticeable as being more reminiscent of Aristotle than of Linnasus.
With regard to generation, a distinction is made between viviparous, ovip-
arous, and gemmate reproduction. In regard to irritability, he differentiates
between insect larvae, worms, and polypi, which have an entirely contractile
or muscular body; vertebrate animals, whose muscles cover the skeleton;
insects and crustaceans, in which the skeleton covers the muscles; and plants,
which possess no free movements. It would be possible, of course, to adduce
weighty detailed objections to this system and its various categories; it
gives evidence, however, of a careful study and a penetrating analysis of
the phenomena of life. And Vicq d'Azyr undeniably possesses a keen eye
for certain manifestations and functions of life — far keener indeed than any
of his predecessors and many of his successors.
In particular he keenly criticizes the current theories of the essence of
304 THE HISTORY OF BIOLOGY
life; on the subject of vitalism, as maintained by Stahl's successors, he holds
that, while it is true that a number of phenomena exist only in living crea-
tures, nothing is gained by referring to the soul as their cause; they should
rather be regarded as physical phenomena and studied through observation
and experiment, but not ascribed to a principle "whereby thought retires
in the belief that everything is done, when in reality everything remains
to be done." This criticism he extends to several current hypotheses; thus,
he rejects the assumption of a fluid controlling the impulses in the nervous
system, for by this particularization of a little-known function a number of
illusions have been created, while such an expression as "nervous force"
would be far more applicable to the actual knowledge we possess of the
phenomenon. Similarly he criticizes the theory current at the time, which
originated in Buffon, that the different parts of the embryo are derived from
corresponding parts in the parents — a theory that even Darwin afterwards
entertained. In disproof of it Vicq d'Azyr adduces the fact that two parents
who have lost one and the same part of the body nevertheless produce normal
offspring. In this point he has thus foreshadowed views that are expressed
by modern students of heredity.
As his teacher in scientific criticism Vicq d'Azyr mentions the philoso-
pher of enlightenment Condillac (1715-80), who enjoyed a great reputation
in his time and who made a special study of the relation of sense-perception
to the consciousness — he asserts that the consciousness is composed of what
the sense-impressions communicate from the outside world — and in con-
nexion therewith he maintained that in science words should exactly convey
the ideas they are intended to denote. By a careful study of his writings
Vicq d'Azyr undoubtedly learnt to realize the necessity for clear ideas and
unambiguous terms in natural science as much as in anything else.
Vicq d' Ax,yr s comparative anatomy
VicQ d'Azyr's influence has been felt not only on account of this criticism,
valuable as it is, but in a still greater degree as a result of his studies in
comparative anatomy, which were unfortunately fragmentary; the prin-
ciples on which he worked, however, he summarized in the form of a pro-
gram for a course of lectures in anatomy and physiology. The subject is first
of all divided up according to the nine life -functions referred to above. Under
the heading "ossification" he first deals in descriptive form with the bone-
structure and its articulation and forms of connexion, then by way of com-
parison the individual bones in different animal forms, and further a number
of physiological experiments in connexion with the growth and regeneration
of bones, and finally he gives an account of the chemical composition of
osseous tissue. Under the heading "irritability" is discussed the muscular
system, first descriptively, then comparatively, as regards both conformation
and finer structure, the vascular and nervous ramifications, and, finally, ex-
Pi
m
1— I
>
p
u
<
u
I— t
eq
Pi
w
I— I
>
<
X
c
2
<
w
1—1
MODERNBIOLOGY 305
perimentally. In the last-mentioned respect are observed both the contraction
of individual muscles when the nerves that are connected with them are
irritated, and the different forms of movement in man and animals — a sub-
ject in which Borelli and Perrault are cited as precursors. Similarly, under
the heading "sensibility" the nervous system is dealt with, followed by
the other functions of the body. The most remarkable point of this exposi-
tion is the detailed comparison made between the same organs in different
animals — a study in which Vicq d'Azyr certainly finds support in the pre-
liminary works of Daubenton; but the former without the least doubt carries
through the comparative program far more thoroughly than the latter.
Its valuable contributions to science
In points of detail Vicq d'Azyr's investigations contain many contributions
of immense value, and also a wealth of ideas of great significance for the
future. He paid special attention to the comparative anatomy of the mam-
mals. He continued and widened the comparison of the bodily structure of
man and the apes which Camper had initiated, arriving, too, at the same
results as the latter (see Part II, p. x6o); on this subject he made a special
study of the musculature of the extremities, and in general closely compared
the extremities throughout the mammalian class. Further, he made investi-
gations into the teeth of the entire vertebrate class; he points out the differ-
ence between teeth fixed in dental sacs and provided with vascular and
nervous systems, and those that are fixed on the jaw-bone; he observes the
dissimilarity in the number and structure of teeth in mammals of different
structure and habits; he draws attention to the pointed teeth of the beast of
prey, the knobby teeth of omnivorous animals, and the enamel-coated teeth
of herbivorous animals; he notes the presence and absence of various kinds
of teeth in different animals. He points out the correlation existing between
different organs in animals; a certain shape of tooth presupposes a certain
type of structure in the extremities and the digestive canal, because all its
bodily parts are adapted to the animal's way of living. He also shows how
these different characteristics give every animal a special role to play in the
great struggle that is constantly going on in nature between the various
life -forms. The weakest point of his comparative investigations is the com-
parison between vertebrates and invertebrates; although he is far more cau-
tious than Geoffroy Saint-Hilaire was later, he nevertheless has no true eye
for the difference between the organs in the important main groups in the
animal kingdom, and, on the whole, it was certainly fortunate for him that
he did not find time to extend his studies to the invertebrate animals. He
even includes the vegetable world in his comparative studies, sometimes
without much success, as when he compares the symmetry in pinnate plants
with that of animals, and sometimes with an insight that looks far ahead
into the future, as when he crosses white and red tulips and finds that the
3o6 THE HISTORY OF BIOLOGY
descendants are white, red, and intermediate. Finally, it is worth mentioning
that as a descriptive anatomist he produced a copiously illustrated work on
the brain and its nervous system — a splendid work for his period and in its
extent the most considerable of all his productions. Thus, in more than one
respect Vicq d'Azyr has left his mark on the history of biology; he will be
especially remembered as a pioneer in the sphere of comparative anatomy.
In Germany at the beginning of the nineteenth century, biology was
dominated by the romantic natural philosophy that was described at the
close of Part II, in which it was also pointed out that during that period
there were working by the side of the natural philosophers a number of
scientists who pursued their inquiries by exact methods, thereby upholding
the traditions of the preceding era and at the same time paving the way for
the succeeding age's magnificent progress in the field of biology. Of these
exact scientists working during a period given over to fantastic dreams some
of the most prominent merit description here.
JoHANN Friedrich Blumenbach was born at Gotha in 1751; his father
was a schoolmaster, and his mother, to whose memory he dedicated a special
work, was a good and gifted woman. Even as a child he was interested in
natural science; one of his keenest delights was putting together skeletons
out of bones that he collected. He studied first at Jena, and then at Got-
tingen, where he took his degree with a dissertation on the human races,
which brought him immediate fame and procured for him, as early as in
the year 1776, the professorship in anatomy at that university. He carried
on his work as a teacher for nearly sixty years, during which he led a quiet
life, interrupted only by a few collecting-expeditions. At last, in 1835, ^^
resigned, and died in 1840. In his old age he was a very original character;
he was regarded as one of the sights of Gottingen, and he was always quite
willing to show himself, especially to distinguished visitors. As an author,
too, Blumenbach is peculiar; his style is heavy and full of long periods, oc-
casionally lightened by dry humour, which is always inoffensive, though
now and then not in the best of taste according to modern standards. It is
also said that his lectures were interspersed with witty remarks, which re-
curred year after year in a given context to the delight of generations of
undergraduates. And Blumenbach had innumerable pupils; he had the ability
both to gather round him and to train scientific experts, and not a few scien-
tists of European reputation derived their knowledge from him. As an author
of text-books and manuals he was very fine for the age in which he lived,
and, generally speaking, he has contributed very largely towards stimulating
his countrymen's interest in the study of nature, which, indeed, was to at-
tain, during the generation that immediately followed his own, unexpected
heights in his country; so that he has honourably deserved the title of
'* Mapsfer Gerfnania," which he enjoyed even in his lifetime.
MODERN BIOLOGY 307
Blwnenhach' s comparative anthropology
As the chief service that Blumenbach has rendered to science is generally
quoted the fact that he introduced into Germany the study of comparative
anatomy — and this at an earlier date than Cuvier introduced it into France.
There can be no doubt that the two collaborators in research BufFon and
Daubentonhave the prior claim to the introduction of comparative anatomy,
but it is certain also that Blumenbach was essentially a comparative anato-
mist and that he brought that science up to a high state of development.
There is especially one branch of it in which he is a pioneer — namely, an-
thropology. Here, it is true, Buffon, with his descriptive and statistical
method, and Camper, with his studies of the facial angle, had paved the
way, but Blumenbach was the first who systematically worked at the sub-
ject, thereby laying the foundations on which all subsequent research has
carried on its constructive work. He instituted a collection of skulls, skele-
tons, and illustrations of human beings of as many different races as he could
procure, and he methodically studied the peculiar characteristics of the ma-
terial he thus got together or was able to borrow from other museums. The
result was a close comparison of the characteristics, both external and in-
ternal, of different human types, and on that basis a division of mankind
into races. A similar attempt had indeed been made before, such as Buffon's,
for instance, but Blumenbach's was the first that really proved successful,
and his five races — Caucasian, Mongolian, Ethiopian, American, and Ma-
layan — have been the foundation on which all subsequent racial divisions
have been based, just as his postulate that the races are varieties of one and
the same species is also regarded as true, in spite of isolated attempts to
create several species. His characteristic descriptions of the cranium, with
accompanying illustrations, have certainly been more recently improved
upon by Anders Retzius, Virchow, Broca, and others, but Blumenbach's
nevertheless form the groundwork on which his successors have built.
Besides the study of races, Blumenbach paid special attention to the
question of determining the characteristics wherein man differs from the
other mammals, and particularly from the manlike apes, whose anatomy he
closely studied. Like Camper, he strongly maintained that man is funda-
mentally unlike the apes; it was he who divided the Linnasan order Primates
into two — Bimana for the human and Quadrumana for the apes. And he
collected as much anatomical, morphological, and psychological evidence as
he could in proof of this. Many of these detailed anatomical characteristics
are without doubt correctly observed, whereas others are the result of ana-
tomical misconceptions, such as the statement that the apes have four hands,
while man has two. This point of view was accepted, however, by the biol-
ogists of the succeeding age, and in actual fact Blumenbach was the origi-
nator of most of the reasons which eventually were to be adduced against
3o8 THE HISTORY OF BIOLOGY
Darwinism by its conservative opponents. In other points, too, these cham-
pions of man's high dignity could be satisfied with Blumenbach's views;
he believes, for instance, in species' having been created as one pair of each —
at least as far as man is concerned — and he holds the view that the Cau-
casian race was the original out of which the others were created later by
"degeneration," due to climatic and economic conditions. During the fight-
ing days of Darwinism, therefore, old Blumenbach received but little
gratitude at the hands of the champions of progress, but a later age must
indubitably do him the justice of acknowledging him as the founder of
comparative anthropology.
Blumenbach, however, extended his research and educational work to
other spheres; actually, like Linnasus, he has dealt with all the three king-
doms of nature. His botanical knowledge is based entirely on Linnasus, whose
system he uses in its entirety; in mineralogy, too, his activities were not
very remarkable. As a zoologist he is likewise limited; he deals with the
invertebrates in a summary fashion and has but little new to tell of them.
The vertebrates, on the other hand, he studied carefully from the comparative
point of view, with special reference to the mammals; he discusses the lat-
ter's anatomy in detail, principally their bone-structure, and his work on
this subject is worthy to be compared with Daubenton's and Vicq d'Azyr's.
In contrast to the last-named he divides his comparative anatomy according
to organs and not according to physiological functions, which gives to the
entire work a far more modern character.
His vitalism
He was also, however, keenly interested in physiological problems and be-
lieved that he had created something essentially new in this field; in a treatise
entitled tjber den Bildungstrieb he expounds a theory of reproduction and em-
bryonic development, which he afterwards advances repeatedly in various
connexions. In this work he first of all gives an account of earlier theories
of evolution; the preformation theory is criticized and rejected, with a cer-
tain degree of satire, his principal target being, as usual. Bonnet's incap-
sulating theory; the spermatozoa are declared to be parasites, and finally
the epigenesis theory is advanced as the true explanation of the phenomenon
of evolution. With C. F. Wolff, he holds that the prospective individual
is evolved out of a completely indifferentiated mass. Blumenbach, however,
rejects his predecessor's theory of " vis essenfialis" (see Part II, p. 2.49), hold-
ing instead that the development is caused by a special "formative force,"
which is displayed not only in the development of the embryo, but also
in all kinds of growth, regeneration, and reproduction in animate beings.
This formative force — nisus jormativus — must not, however, be confused
with other "life-forces," such as irritability and sensibility, but it operates
with those forces in order to maintain life. Blumenbach specially points out
MODERN BIOLOGY 309
that these "forces" are merely expressions by which to denote phenomena,
the cause of which we do not know, but the effects of which we can observe.
By the side of this, however, he speaks also of the body-mechanism. Thus
we find that Blumenbach did not create, like Stahl, any elaborated vitalistic
thought-system, and his speculation in general is not particularly logical.
His service lies rather in the field of comparative observation than in that
of speculation.
Blumenbach's contemporary and equal in scientific reputation was Sam-
uel Thomas Sommerring (175 5-1830). He was born in the Polish town of
Thorn, though of a German family; his father was town physician, and the
son was destined early for the medical profession. He carried out his elemen-
tary studies at Gottingen, where Blumenbach was one of his younger teachers;
he afterwards studied anatomy, in Holland under Camper and in England
under Hunter. His youth recalls that of Swammerdam, in so far as his father
desired to see him early established in practice and stubbornly opposed his
going in for expensive scientific studies; but the young man refused to give
in; through his personal ability and the mediation of friends he managed to
secure extended help towards his studies away from home and thus continued
his scientific work in difficult economic circumstances, until he was able to
earn not only a good reputation, but also his daily bread. He held profes-
sorships, first at Kassel, then at Mainz and Munich; but during that period
spent some years as a practitioner at Frankfurt am Main, where he married
and found his true home. There, too, he spent the last ten years of his life
in peace and happiness, surrounded by friends and continuing his scientific
work until the end.
In his general conception of nature Sommerring was to a certain extent
influenced by the speculations in mystical natural philosophy in which his
age indulged. At Kassel he entered the Rosicrucian Brotherhood and in it
carried on both alchemy and spiritualism, although he afterwards realized
his delusion in both respects. At Mainz, where his activities were most pro-
ductive, he applied himself exclusively to anatomy and carried out in this field
a number of valuable investigations on special subjects. Like Albinus, whom
he chose as his model, he employed a clever draughtsman and with his assis-
tance published several excellent compilations, one on the human body in its
entirety — which was never completed, but the published sections of which
are remarkable for their clear method of presentation and their sound de-
scriptions — as well as a number of special investigations, such as a treatise
on monstrosities of various kinds, which he spent some years in collecting,
and, further, a comparative study of the visual and auditory organs in differ-
ent races of mankind, and finally some investigations on various subjects
and of varying value.
3IO THE HISTORY OF BIOLOGY
Sommerring s work on the brain
His curious work fiber das Organ der Seele, which he dedicated to Kant, is a
combination of anatomical inquiry and philosophical speculation. In it he
gives a detailed description of the brain and its nerves, illustrated with
splendid engravings. His account of the origin of the nerve-stems in particu-
lar is admirable, considering the age in which it was written, and he has
rendered a still greater service to science by treating, for the first time, the
sympathetic nervous system as a pair of nerves independent of the central
nerve system, "which pair is in indirect, but not direct, connexion with the
brain and spinal cord." The whole of this study of the brain, however, forms
the basis of a highly fantastic speculation upon the brain as the organ of the
soul, or, to be more exact, upon the location of the "sensorium commune,"
which in the German is translated as "das gemeinschajtliche Emffindungsort."
By this is meant that part of the brain in which the sense-impressions con-
verge and co-operate. Ideas of this kind in regard to the localization of the
soul in the brain had indeed long been current; Descartes adopted for this
purpose the glandula pinealis, Perrault the ??2edulla oblongata; Swedenborg
alone was guided on the right path by his brilliant intuition when he drew
attention to the pyramid-cells of the cerebrum. Sommerring tries to prove
that all the cerebral nerves open into the central cavity of the brain and that
in connexion therewith the cerebral fluid is the organ of consciousness; the
only point that worries him is: "Kann eine Fltissigkeit animirt sein?" which,
however, he answers in the affirmative on arguments derived from the Bible,
Aristotle, and modern writings. This assertion, which in our own day, when
protoplasm and its derivatives have so often had to serve as wholly or at
least partially fluid, should not be regarded as utterly absurd, nevertheless
aroused grave doubts in the minds of Sommerring's contemporaries, just as
his philosophical argumentation in general shocked the students of natural
science; his good friend Goethe wrote him a letter in which, with reflective
and observant criticism, which he unhappily did not always employ in his
own writings, he warns him against letting philosophical speculations inter-
fere in scientific investigations. And Sommerring in actual fact learnt wis-
dom from the opposition he met with; his later works are in the main based
on exact natural science; his reputation as one of the leading anatomists of
his age was greatly enhanced by them and has been confirmed by posterity.
There was one scientist, a contemporary of Sommerring's, who ap-
proached far more nearly to the modern conception of the structure of the
brain and nervous system, but through his own fault he managed to acquire
a somewhat doubtful reputation; this was the "phrenologist," Franz Joseph
Gall (1758-1818). Born at Baden, he went as a medical student to Vienna,
became a doctor, and carried on a medical practice there. At the same time
he began to interest himself in the study of the brain's structure and its
MODERN BIOLOGY 31I
manifestations of life. He promulgated his ideas on the subject both in
writings and in public lectures; when these were prohibited as being "ma-
terialistic," he left Vienna (in 1805) and wandered about Germany for a
couple of years, accompanied by his friend and pupil Spurzheim, everywhere
demonstrating his ideas and wherever he went attracting the attention of the
public, which was occasionally flattering, but often quite the opposite.
Eventually he settled down in Paris, became naturalized, and lived on a
practice which had its peculiar features — for instance, he kept strictly
secret the composition of the medicines he prescribed — and which brought
him into strained relations with other doctors. He was also on bad terms
with scientific specialists; universities and academies closed their doors to
him. The public became all the more interested in his doctrines, which were
promulgated after his death by many, mostly dilettanti, who brought his
theories into utter discredit, so that ultimately they were entirely forgotten.
Gall's theory of brain and nerves
Nevertheless, Gail has exercised an undeniable influence even upon serious
science. For he was without doubt one of the most brilliant brain-anatomists
of his age, and the ideas he produced on the subject have proved of great
significance for the development of that branch of science. In his exposition
of the nervous system he does not start from the brain, as his contemporaries
did, but from the simple nerve-fibre, which he considers to be the simplest
type of nerve; it is found even in worms, and out of it "nature has evolved"
all the higher nerve-forms: the ganglia as the junctions of several nerve-fibres,
and the spinal cord, which consists of a series of ganglia drawn through by
a mass of nerve-fibres and connected by means of cross-fibres. Through the
spinal cord the nerves lead up to the brain, where they end in the cortex-
substance which represents the brain's "ganglion"; in this latter are com-
bined the functions of the nervous system, particularly in the folds of the
cerebrum; this, indeed, is the reason why, the more highly the great brain is
developed, the greater is the intelligence. In the cortex of the brain are situ-
ated the different intellectual qualities of man; these qualities are due to he-
reditary tendencies and together form the soul, which is thus not confined to
any particular spot in the brain, as earlier anatomists had declared. What is
new and of value to the future in this nerve theory is, first of all, the emphasis
he lays on the significance of the nerve-tracts; further, and above all, the
placing of the soul-functions in the cortex of the great brain; and, finally, the
assumption of hereditary intellectual tendencies. In particular, the idea that
the cortex of the great brain is the organ of intelligence has been fully veri-
fied. It is not easy to determine how far Gall, who undoubtedly possessed a
thorough knowledge of the details of cerebral anatomy, himself established
this fact, or how much he borrowed from his predecessors. It is at any rate
peculiar to him that he cites among authorities on this subject even
312. THE HISTORY OF BIOLOGY
Swedenborg, whom he must thus have studied, and who may well have been
able to inspire him with ideas pointing in that direction; at all events, he
has not adopted his predecessor's theory of the essential part played by the
pyramid-cells in the work of the brain; on the contrary, he overlooks them
and believes that the cerebral cortex is composed of matted nerve-fibres.
Gall's theory nervertheless represents a great advance towards the modern
standpoint and has undoubtedly exercised considerable influence on the
development of cerebral research, in spite of contemporary opposition. This
is also the case with Gall's assumption of hereditary intellectual tendencies,
which represented a definite advance in face of the naive belief of the philoso-
phers of enlightenment that all men possess like tendencies to virtue and
genius, which require only proper education in order to be able to develop.
Unfortunately Gall went to the most ridiculous extremes in developing his
theory; he sought and discovered in the brain organs for all kinds of intellec-
tual and moral qualities, for genius and beauty, love and piety, and even for
stealing and murder. And he went so far as to imagine that he could discern
these very qualities in the irregularities on the surface of the skull, since he
believed the skull to be exactly fashioned after the brain. This study, which
he called "cranioscopy," rapidly degenerated into sheer humbug; in particu-
lar, the discovery of the "bumps of genius," which were supposed to denote
special talent, brought much profit to quacks and rogues. This expression
has survived in modern phraseology as the best-known relic of Gall's activi-
ties, which has served to conceal the really sound work that he accomplished
in biological science.
Another scientist who was closely connected with natural philosophy
was JoHANN Christian Reil (1759-1813). Son of a clergyman of East Fries-
land, he studied medicine at Gottingen and Halle, practised for some years
in his home district, was then appointed professor of internal medicine at
Halle, and at the same time became town physician there. When the Uni-
versity of Berlin was founded, he was elected a professor, but resigned his
post upon the outbreak of the War of Independence against Napoleon and
volunteered as an army surgeon. When acting in that capacity he fell a
victim to the typhus epidemic that raged during the war.
Reil's life-theory
Reil's influence has been both many-sided and important. He was highly
esteemed by his contemporaries; among Scandinavian doctors Israel Hwasser
in particular studied and admired him. As a practitioner he enjoyed a wide
field of activities; he recorded in a compendious work all the knowledge that
his age possessed of fevers and their treatment. Still more influential, how-
ever, was his work in the sphere of psychiatry, which he radically reformed;
he effected improvements in the appalling conditions prevailing in the luna-
tic asylums and insisted upon the elevation of psychiatry to the position of
MODERN BIOLOGY 313
an independent branch of study at the universities. All his practical endeav-
ours, how^ever, he preferred to base upon a careful study of the functions of
the body; for this purpose he started in the year 1796 the journal Archiv fur
Physiologic, which under various names and editorial conditions has sun
vived up to the present day. In an essay in this journal he expounded a
general biological theory that exercised great influence on his own time and
is therefore worthy of reference. This essay, "Von der Lebenskrajt ," contains,
like many others written in that age of pioneers, a number of ideas fruitful
both for the contemporary world and for posterity, side by side with a mass
of uncritical and fantastic nonsense. After a philosophical introduction
touching the terms "matter," "phenomenon," and "idea," Reil criticizes
the vitalistic speculation of preceding ages and declares, with an obvious
reference to Stahl, that phenomena in the animal kingdom cannot emanate
from an immaterial soul, because assumptions as to supernatural influences
explain nothing. Rather, the basis of all phenomena in the animal body that
are not ideas must be sought only in corporeal matter and in "the form and
composition" in its various constituent parts; to matter's different "compo-
sition" in muscles, nerves, and bones are due the different properties and
functions of those parts. Reil has here learnt not only from Stahl, but also
from the animal chemistry that started in connexion with Lavoisier and had
been developed in his own time. Unfortunately, however, he does not by
any means come up to the standard already reached by his contemporaries
in this respect; thus, he did not realize the relation between the interchange
of gas in plants and animals, a fact which Schelling, for instance, during the
same period realized well enough to be able to ascribe fundamental impor-
tance to it (Part II, p. xyy), and he shares the ancient popular belief that
the grain of seed in the earth and the still unbrooded tgg are "dead" and ac-
quire life by being provided with warmth and other fine components of life.
Herein lies the weakness of Reil's speculation, and, generally speaking, both
chemistry and philosophy lead him into somewhat strange paths. Having
thus started by defining the idea of force in nature and the relation between
phenomena and the properties in matter which produce them, he goes on to
explain the life-force in animate creatures as being the relation between
more individualized phenomena and a special kind of matter, wherein a
differentiation is made between vegetative force in plants, animal force in
animals, and reason in man. Growth in inanimate and animate nature is
declared to be of an identical character, so that animal growth is at once
termed ''tierische Kristallisafion." But besides these fanciful ideas Reil suc-
ceeds in producing a definition of the term "organ" itself that undoubtedly
represents a real advance. Starting from the elder Darwin's theory of fibre
as a basic component in the animal organism (Part II, p. X95), he describes
various categories of fibres: cell-tissue fibres, bone-, muscle-, and nerve-fibres.
314 THE HISTORY OF BIOLOGY
Out of these are formed more complex organs, such as nerves, bones, liga-
ments, cartilages, muscular mass; and finally there are formed of these com-
ponents in various proportions the higher organs — ■ namely, intestines,
sensory organs, and musculature. Here we undoubtedly catch a glimpse of
the idea of tissue, which Bichat afterwards developed, independently of Reil
and more universally and radically than the latter. There is no doubt, how-
ever, that Reil also helped the succeeding generation, in Germany especially,
to define the terms on which anatomical science has since developed.
2.. Humboldt
In this connexion there is also worthy of mention a scientist who, like
those described in the foregoing, belongs both to the history of natural phi-
losophy and to that of exact natural science, but whose fame far outshone the
rest and who is universally looked upon as one of the greatest personalities
in the whole range of science: Alexander von Humboldt. He was born at
Berlin in 1769 of a distinguished and wealthy family; his father was chamber-
lain at the court, his mother came from a French family, who had gone into
exile for their Protestant faith. Having studied at the University of Gottingen
and at the mining academy at Freiberg, he entered the service of the Prussian
Mining Department and worked there for some years, until an ample in-
heritance placed him in a position of being able to devote himself to natural
science without having to earn his living. After preliminary studies and
travelling in Europe, he equipped at his own expense in the year 1799 a
journey of exploration to South America, which region he traversed in
various directions and explored so thoroughly that he was called the second
discoverer of America. After five years out there he returned home with rich
collections, which it took many years to work up. For this purpose he spent
a long time in Paris and there published an extensive account of his expedi-
tion, which made him world-famous. He spent all his fortune on the journey
and its description, but the King of Prussia indemnified him by presenting
him with a well-paid post as chamberlain; he rejected offers of university ap-
pointments. In 1817 he settled in Berlin and there spent the rest of his days,
except for a short expedition to Russia and Siberia. In close contact with the
royal family, yet retaining the liberal ideas of his youth, respected as one
of the great men of science and highly esteemed for his lovable person-
ality, he lived to a great age, working incessantly at different branches of
science, though certainly towards the end with diminished energies. He died
in 1859.
Humboldt was an unusually highly gifted personality, artistically as
well as scientifically, and he has exercised an extraordinarily varied influence
MODERN BIOLOGY 315
Upon the development of natural science, although he did not rise to the
highest levels in any particular sphere. As a scientific explorer he is without
a rival and he raised geography to the rank of a science. Climatology es-
pecially owes its fundamental principles to him: thus, the method of indicat-
ing on the map by means of isothermal lines places having a similar annual
temperature was invented by him. He devoted many years of methodical
study to terrestrial magnetism, and the magnetic-meteorological observa-
tories which are now established throughout the globe have him to thank
for their existence. As a geologist he deserves especially well of science,
owing to his studies of the problem of Vulcanism; he established the fact
that the volcanoes exist grouped in ranges along cracks in the earth's crust.
But he was also highly interested in biological problems.
Humboldt' s idea of life- force
In his youth he expounded a theory of life as a whole in the form — charac-
teristic of the man himself and of his age — of a mythological story en-
titled Die Lebenskraft oder der rhodische Genius. The gist of it is that life is
maintained by a force that prevents the elements of which the body is com-
posed from obeying the laws of affinity that hold good in inorganic nature.'
In his old age, however, he abandoned this fantastic theory and in his later
writings utters a warning against any kind of speculating upon the life-
force. He displayed greater exactitude, however, in his investigations,
published shortly before his South American expedition, into the influence of
electricity upon muscles and nerves, which he carried out partly with himself
as subject, and which, together with a number of natural-philosophical
illusions, contain ideas that have been utilized in research work of a later
period in connexion with electrical phenomena in the animal kingdom.
His vegetable geography
The greatest service rendered to biology by Humboldt, however, was his
creation of vegetable geography. Even as early as in Linnasus we found a
lively interest and a keen eye for the life-habits of plants. Linnasus's investi-
gations into the question of the habitat and distribution of plants (Part II,
p. 115) were, however, based entirely on his classification system. Humboldt's
interest in plant life, on the other hand, is at the very outset of quite a
different nature. As is the part of a natural philosopher, he takes as his
starting-point life in its entirety, examines its various manifestations, and
finally dwells on the special advantages which soil and climatic conditions
offer to the vegetable world in different latitudes. He puts the question:
How is the shape of plants affected by these conditions of life? And he
searches for the connexion between the impression made by the landscape
^ The idea contained in this story is without doubt directly or indirectly influenced by
Stahl's previously mentioned theory of the soul as the force that prevents the chemical com-
ponents of the body from disintegrating (see Part II, p. 181).
3l6 THE HISTORY OF BIOLOGY
on the observer and the shape of the plants that dominate the landscape.
Thus he produces semi-artistic, semi-scientific pictures of vegetation in
different latitudes; he declares that each latitude possesses its own charac-
teristic natural physiognomy, and he finally differentiates between certain
vegetable types, not according to systematic characters, but according to the
impression the observer receives of their form as a whole. He distinguishes
sixteen of these landscape-forming vegetable types, though he states that
their number could certainly be increased. Among these types may be men-
tioned: the palm type, the banana shape, the heather type, the cactus type,
the orchid type, the fir type, grasses, ferns, lilies. The whole of this concep-
tion of plant life and this grouping of its individual components according
to common conditions of life, instead of according to the nomenclature
of species, represent a new idea; it is true that here Humboldt has learnt
something, as he himself acknowledges, from BufFon, as well as from a
number of earlier describers of landscapes, but out of these ideas and as the
result of his own observations he created a new field for research, which was
cultivated and extended at a later period with great success.
His cosmos
During the last decades of his life Humboldt devoted himself to formulating
a universal cosmology, which was intended to reproduce every imaginable
conception of and all the known facts about the universe : the purely scientific,
the historical, and the artistic. This gigantic work, the execution of which
was far beyond the powers of one single man, he called Kosmos; its first part
was published in his seventy-fifth year and a final part of the unfinished work
came out after his death. Never has any natural scientist of modern times
conceived a plan on a grander scale, and though its execution is naturally
both fragmentary and defective, the work nevertheless contains a vast
amount of valuable material in the way of facts and is, besides, like all
Humboldt's work, unequalled in style. Romantic natural philosophy's idea
of a uniform conception of nature has received in Humboldt's Kosmos its
most glorious memorial; it seems almost symbolical that its creator should
have died in the same year as that in which Darwin published his work on
the origin of species; the modern theory of evolution stepped in where
natural philosophy ended.
3 . Lamarck
Jean Baptiste Pierre Antoine de Monet, usually styled Chevalier de
Lamarck, was born in Picardy , in northern France, in 1744, ^^^ ^^ ^^^ young-
est of a large and poor noble family. At an early age he was sent to a Jesuit
school with a view to eventually securing a comfortable living as a priest.
MODERN BIOLOGY 317
From the outset this prospect failed to have the least attraction for him, but
as long as his father lived, he had to obey him. When he was seventeen,
however, his father died, and he inherited a sum just sufficient to enable him
to buy a nag; on this he rode away and joined the French Army, which at
that time was in the field during the Seven Years' War. On the day after he
enlisted, a battle took place, in which his company suffered severely, losing
all its officers and non-commissioned officers, whereupon Lamarck, with his
one day's war-experience, collected the survivors and held out at his post until
help arrived. This deed was rewarded with a lieutenant's commission, but
his promotion went no further; he was sent to Toulon on garrison duty, and
on the conclusion of peace he resigned his commission for reasons of ill
health and was granted a small pension. He now had to look about him for a
fresh means of livelihood, and for this purpose betook himself to Paris;
there he remained for the next fifteen years as a literary hack, living in a
garret in the Quartier Latin just the kind of Bohemian life that has so often
been described in novels. During these difficult years, however, there de-
veloped in Lamarck an ever-increasing love of natural science, particularly
botany; even during his garrison life on the shores of the Mediterranean the
abundant and wonderful flora of that coast had deeply interested him, and
this love of knowledge grew apace in Paris, where in those days the interest
in animate nature was kept alive by Buffon. It was he, too, who paved the
way for the scientific success of the penniless writer; he became interested
in a flora of France that Lamarck had written and procured his admittance
to the Academy of Science. Further, Lamarck was commissioned to travel
through several European countries as companion to Buffon's young son,
and he finally became an assistant in the botanical department of the natural-
history museum. It was during the Revolution, however, that Lamarck
first obtained a secure position; the National Convention, which wanted to
reform everything, instituted a number of professorships, including two in
zoology. As no more suitable candidates could be found, the one chair was
offered to the botanist Lamarck, and the other to Geoffroy Saint-Hilaire,
who had till then been mostly occupied with mineralogy. These two im-
provised zoologists shared between them the duty of lecturing, Geoffroy
undertaking the vertebrates, and Lamarck the invertebrates. Thus, at the
age of fifty Lamarck started research work in the field in which he was
eventually to win fame as a pioneer. The rest of his life passed in assiduous
work in the career he entered so late; retiring and modest as he was, he
sought no outward honours, nor did he win any; he remained throughout
his life in poor circumstances, especially at the end, having lost by unsuccess-
ful speculation what little capital he had saved. He suffered, too, from domes-
tic troubles more than most people; he was married four times and lived to
see all his unions dissolved by death, while of his seven children the majority
3l8 THE HISTORY OF BIOLOGY
also died prematurely. Two daughters, who devoted themselves entirely to
administering to him, were his one consolation in his old age; with their
aid he was able to carry on his work unremittingly to the end, although he
was blind during the last years of his life. He died in 18x9, and a year later
the last part of the work that had occupied him up to the last was published.
Just as the strangeness of Lamarck's fate is unique in the annals of biol-
ogy — a discharged lieutenant without any scientific grounding, who from
being a Bohemian literary hack works himself up to lasting fame as a scien-
tist and who at the age of fifty becomes professor in a subject that he had never
studied before — so his posthumous reputation has likewise been unique.
By his contemporaries he w^as mainly looked upon as a systematist, and, as
we shall find later on, he certainly did accomplish valuable and sound work
as one. But besides that he published a number of works on evolutionary his-
tory based upon speculation; these attracted little attention, however, either
in his own day or in the immediately succeeding period. They were neglected
by the natural-philosophical school for reasons that will be explained later,
and were regarded by the subsequent representatives of exact research as
fantastic speculations. It was not until after the launching of the modern
theory of the origin of species that Lamarck came into his own. Haeckel in
particular, who searched everywhere for precursors of that theory, the
promulgation of which he made his mission in life, referred to Lamarck as
a pioneer of modern natural research, and there followed in his footsteps
a whole group of scientists who saw in Lamarck's theories the basis for a
correct view of evolution in nature. During the last few decades this so-called
neo-Lamarckian school has, it is true, fallen off considerably in both num-
bers and influence, but Lamarck himself is still counted one of the pioneers
of modern biology.
Lamarck' s multifarious ivorks
The cause of these varying opinions lies essentially in the very character of
Lamarck's scientific productions. As will be seen from the above account of
his life, he was a self-taught man, without any systematic scientific training,
with the result that his production to a large extent bears the mark of dilet-
tantism — many-sided interests, vagueness of both thought and expression,
daringly brilliant ideas side by side with foolish fancies. His earlier works
especially — up to about the close of the century — are extremely multi-
farious as to contents, as well as of unequal value. Besides a number of partly
still valuable botanical writings, he wrote numerous works on meteorology
and geology, as well as a collection of essays with the striking title of
Memoires de physique et d' his f aire nafurelle, etablis sur des bases de raisonnement
independantes de toute theorie. Towards the close of his life, however, he con-
centrated entirely upon zoology and in that field produced his best works. His
enthusiastic admirers have as a rule passed over his earlier speculations in
MODERN BIOLOGY 319
silence, yet without a knowledge of them it is impossible to gain any idea of
Lamarck's scientific development, all the more so as throughout his life he
firmly adhered in all essentials to the views he held in his youth.
In the above-mentioned work, Memoires de physique, Lamarck endeav-
oured to form a general theory of existence, a combination of physics,
chemistry, and physiology. This theory represents a continuous attack upon
what he calls "pneumatic chemistry" — that is, Lavoisier's quantitative
method (Part II, ch. xii). For Lavoisier himself Lamarck has nothing but
praise, his polemics being invariably objective and honest, but on the com-
position of things he has ideas that are entirely his own. Lavoisier had con-
ceived combustion as a process of oxidization; Lamarck finds this explanation
absurd — the idea of oxygen's being an essential component of both water
and air is in his opinion utterly irrational; no chemist has ever seen it and
nobody has been able to prove its actual existence. And equally irrational is
the theory of chemical affinity as a cause of chemical associations between
the elements: "It is not compatible with reason and is therefore impossible."
As essential components of nature Lamarck assumes the four known ele-
ments — • fire, air, water, and earth — and adds a fifth, light. The purest
earth is — rock-crystal. The chemical associations are not at all bound to-
gether by any affinity; rather, they strive to disintegrate into their simple
components. What creates chemical associations on the earth is exclusively
life; all inorganic associations that exist — rocks, minerals, metals — are
disintegrated remains of living beings. Lamarck sets up an evolutionary series
that is unique of its kind, beginning with blood, bile, urine, bone-substance,
snail-shell, and proceeding to increasingly greater "disintegrations" through
shell-lime, marble, gypsum, to precious stones, metals, and lastly "simple"
rock-crystal. The problem of what life really is is of course a question that
largely occupies the mind of Lamarck and is discussed by him with great
particularity. The essential factor in life he finds to be motion; an animate
being is composed of various parts which affect one another and are kept in
motion partly by mutual influence and partly by influence from without, and
it undergoes constant change in consequence of this motion. Life itself is
motion and nothing else — that is, a purely mechanical phenomenon. The
essential components in the living body are partly solid (fibres and mem-
branes), partly liquid (blood, lymph, and other special "fluids," of which
more later on). Of the functions of life, secretion within the organism is an
expression for the afore-mentioned efl"orts made by the chemical associations
to disintegrate; nutrition counteracts these eff'orts by providing the living
being with fresh substances, a difference being made between the power of a
plant to form out of simple alimental substances complex bodies, and the
dependence of animals upon these same complex products for their nourish-
ment. — In these and other phenomena, in both animate and inanimate
310 THE HISTORY OF BIOLOGY
nature, there is, according to Lamarck, incorporated as an essential com-
ponent fire, which penetrates the whole of existence; it is a "fluid," which
appears under various modifications, as heat, as electric and magnetic fluid,
and in living beings still further specialized. It produces colour-perceptions,
sound — Lamarck denies that the air conveys the sound, for a cannon-shot
is heard at a distance better with the ear to the ground than in the air — ■
, and, further, chemical changes, the various kinds of which it would take too
long to enumerate here. The serious offences against contemporary physical
and chemical knowledge of which Lamarck is guilty in this work will have
been sufficiently illustrated by the above. And it would be a waste of time to
trace the sources out of which he created these wild fancies; he himself, in-
deed, asserts that they are "independent of any theory" and they are charac-
terized from beginning to end by sheer dilettantism. Fortunately, however,
Lamarck did not retain this standpoint always; although more than fifty
years old when he published his Memoires, he managed to escape out of the
helpless maze of thought that they involve and to create works which have
kept his memory alive even up to the most recent times — a spiritual test of
strength indeed, which is almost without its counterpart in the history of
science. That this was so is not due to his having acquired any essentially
better knowledge of physics and chemistry than others,^ but to his having
applied himself whole-heartedly to zoology. In this field, thanks to his long
experience as a lecturer and a museum-worker, he had gained a many-sided
knowledge of form, whereon he was able to base a system of thought that
was not only original, but also truly scientific, as regards both form and sub-
stance.
The result of Lamarck's theoretical speculations in the sphere of bi-
ology — he it was, in fact, who created the word "biology" — is recorded
in three separate works: Kecberches sur V organisation des corps vivants, of i8ox;
Philosophic ^(^oologique, of 1809; and the introduction to his great work His-
toire naturelle des animaux sans verfebres (i8i5-xx). The first of these presents
in short and concise form the theory of the development of life that made
Lamarck famous. It has been completely overshadowed, however, in the
history of biology by Philosophie Zfiologique, which is the one work of
Lamarck that is regarded as a classic and which has in more recent times been
frequently reprinted and translated into many languages. It is really an ex-
pansion of the previous work, full of repetitions and containing a number
of additions, which in many instances, but not in all, are improvements.
In the third of these works the author once more recapitulates the theory in
summary form, as he entertained it towards the close of his life.
2 In his latest works, it is true, he acknowledges the existence of oxygen, a fact of which
he had apparently been convinced by some chemist, but on the whole he maintains the old stand-
point.
MODERN BIOLOGY 3x1
His life-theory: life is motion
Lamarck begins his work Kecherches with a protest against that dry systema-
tization that is content with differentiating as many species as possible with-
out troubling to make a comprehensive survey of the connexion between the
life-forms in nature. Rather, he would start by regarding life in its entirety,
and he thereupon finds, in accordance with his conception referred to above,
that the most essential quality of life is motion. All that occurs in life is
motion; through it the organism strives to develop and to specialize the
organs; motion is also the absorption of nutriment, whereby the individual,
during its days of physical power, compensates for the losses caused by excre-
tion, whereas during the later period of life excretion becomes superior to the
power of absorbing nutriment, so that eventually death results; it is through
motion that development proceeds in every living being, the fluids of the
body making their way through the surrounding solid parts, with the result
that in these latter are formed organs which assume various functions, and
canals which convey nourishment to them.^ Thus is gradually formed not
only the individual, but also, step by step, all living beings of various types,
while the qualities that have been developed in the individual life-forms are
transferred by reproduction to the descendants. On this basis it is possible to
place all living beings in one series, beginning with the lowest and ending
with the highest. It is more instructive, however, to examine the organiza-
tion of animals in the opposite direction, in that, if we start from the highest
forms, we can follow the "degradation" that appears in the series, one organ
after another becoming changed, simplified, and finally disappearing. The
mammals are naturally the highest; they are the only creatures that really
produce their young alive; they possess milk-secretion, independent lungs,
and complete diaphragm. The birds come lower than the mammals, for they
lay eggs, their lungs are fixed, and they have no diaphragm. Below these two
warm-blooded animal groups come the reptiles, owing to their cold blood and
incompletely formed heart and lungs, which latter are in certain forms repre-
sented during earlier stages by gills (the batrachians, as is well known, were
still at that time grouped together with the reptiles); further, the two pairs
of extremities in these animals gradually disappear, wherefore the snakes,
which possess no extremities, are the lowest of the order of reptiles. Upwards,
again, the transition between reptiles and birds is formed by the Chelonia
(tortoises), while the then newly-discovered duck-billed platypus assumes
the same role between birds and mammals. The fishes, on the other hand, are
lower than the reptiles, for they have entirely lost lungs and extremities;
that is to say, their fins are not real extremities. With the transition from the
fishes downwards the backbone and the inner skeleton disappear from the
^ This theory recalls a similar one of Caspar Friedrich Wolff's (Part II, p. 2.50), but it is
uncertain whether Lamarck knew his works — at any rate, he never quotes them.
3X1 THE HISTORY OF BIOLOGY
animal kingdom. Of the invertebrates, the molluscs stand highest, for they
have gills like the fishes and possess brain, nerves, and single-chambered
heart. Next to them come the Annelida, which Lamarck, after Cuvier, dis-
tinguishes from the worms, and which he has named; they likewise breathe
by means of gills, sometimes visible, sometimes concealed in the skin; more-
over, they possess a nervous system, a vascular system with red blood, and a
pair of extensions thereof corresponding to the heart. These are followed by
the crustaceans, also possessing gills and heart, but after them these latter
organs disappear from the animal kingdom. The spiders come next; because
they have a concentrated respiratory system and emerge from the egg in the
same form as they retain afterwards, they are above the insects, which pos-
sess scattered tracheae and undergo metamorphosis. With these animals, in
Lamarck's view, sexual reproduction disappears from the animal kingdom.
Thus, the worms, which follow next in the series, are reproduced by gemma-
tion; as a matter of fact, they may possess a nervous system and tracheas.
With them disappear visual organs and nervous system from the animal king-
dom. The next class is the Radiata, another systematic creation of Lamarck's;
these animals lack visual organs, but possess organs of generation — though
sexless — ■ and are thereby distinguished from the polypi, which possess no
organs at all.
Lamarck having thus classified the animals in a series on a basis of the
absence or presence of certain principal organs, he goes on to state that the
sequence thus formed does not refer to the separate animal individuals, but
to the great masses of animals that form one entire class; within such a class
it is possible that, owing to dissimilarities in less essential organs, ramifica-
tions may take place in various directions, but the above arrangement, which
has been made in the animal classes on the basis of the structure of the most
vital organs, is presented with such certainty that "no enlightened natural
scientist will be able to produce another." It shows how, the higher we come
in the series, the greater becomes the specialization in the organs, while the
lower we go, the simpler we find the organs becoming and the wider their
functions. On this ever-increasing specialization of the organs Lamarck now
bases his theory of how the various life-forms have arisen, a theory which at
the very outset he formulates as follows: "It is not the organs — that is to
say, the form and character of the animal's bodily parts — that have given
rise to its habits and peculiar properties, but, on the contrary, it is its habits
and manner of life and the conditions in which its ancestors lived that has
in the course of time fashioned its bodily form, its organs, and its qualities."
He seeks to prove this basic argument by innumerable examples: moles and
blind mice have lost their sight as a result of living underground for several
generations, the ant-bear its teeth through swallowing its food whole;
waders have acquired long legs and long neck through stretching those parts
MODERN BIOLOGY 313
of the body in their search for food on the shores, swimming birds their
webbed feet through stretching out their toes during their movements in the
water. In connexion herewith he declares that if a number of children were
to be deprived at birth of their left eye and they were allowed to have children
by one another, there would eventually arise after a few generations a one-
eyed race of men. But it is not merely influences of this detailed kind that
re create the life-forms; but also the effect of geographical conditions in
general; climate, humidity, abundance or scarcity of food, have in the long
run had a transforming influence upon animals as a whole and have produced
new organs or caused the old ones to disappear. This is rendered possible
owing to the fact that there is an infinity of time at the disposal of evolution;
"Time has no limits," says Lamarck explicitly. Thus life becomes purely
and simply a mechanical process; all its manifestations are motion and noth-
ing else. It is true, existence has a Supreme Originator, for whose name
Lamarck always si^ows respect, but His greatness lies in the fact that He has
created nature in such a way that it has developed its profuse multiplicity
without interference from without. This development out of given qualifica-
tions Lamarck does not succeed in establishing, however; "nature" appears
constantly as a creative power and is spoken of in terms suggestive of a
personality; this is especially so in Philosophie ■::oologique, from which a few
examples will be cited further on in this chapter.
His theory of life- fluid
In the continuation of Kecherches Lamarck further develops his mechanical
theory of life, which to him is really the main problem, in which his evolution
theory is only one detail among many. To his mind, life itself is the condition
in all the parts of the body that makes their organic movements possible.
This condition consists in the existence of " T or gasme vital," a state of tension,
a "tonus," which maintains the molecules in the soft parts of the body in a
definite position and which by increasing and diminishing enables the organs
to contract and expand. The cause of this tension is a fluid that is secreted
by the blood and thence absorbed by all the organs of the body, but is particu-
larly concentrated in the nervous system. This fluid is really a peculiar variety
of fire, related to heat and electricity, a "feu ethere." It is transmitted on
fertilization from the male sexual product to the embryo, which derives
life from it — immediately in mammals, but not until later in the bird's ^gg,
which only receives life by brooding. But this same fluid exists scattered
everywhere throughout nature, so that everywhere, and especially in hot
countries, with their humid climate, there takes place a spontaneous produc-
tion of life. Lamarck asserts that this spontaneous generation under the influ-
ence of heat, light, and electricity goes on incessantly, the lowest animal
forms — and even plant forms — being continually reproduced out of
inanimate matter; he declares it to be probable that the fresh-water polypi
32-4 THE HISTORY OF BIOLOGY
freeze to death every winter and spontaneously generate again every spring.
According to his idea, the most primitive living creatures consist of a mass of
gelatinous substance, which absorbs nourishment through pores on its sur-
face. Out of these nature gradually evolves a special organ for the admission
of food; first there arises as a result of the movements of the animal a small
depression in which the food can easily collect; through the pressure ex-
ercised by the food, this slight hollow expands into a sack-like cavity, which
similarly becomes in process of time still further extended; thus arose the
polypus's digestive canal. The next important life-property which nature
developed was reproduction; this consists in reality of a growth over and
above the normal dimensions; a division must therefore take place, and
actually does so in the lowest animals, the Infusoria, which never die of old
age, but divide themselves in two when they have attained a certain size.
Through the division's not being uniform, gemmation arises, which is the
manner of propagation characteristic of the polypi; when this is repeated, one
particular area becomes specialized for the purpose, and thus originated the
internal gemmation by means of which the Radiata propagate. Through
further evolution in this direction there arose the eggs, being incomplete
buds which, in order that they may develop further, require to be influenced
by the male sexual product.
The evolution of man
Here Lamarck interrupts his exposition of the origin of the most vital organs
and proceeds direct to a consideration of the evolution of man. Like Camper,
he emphasizes the differences between the anatomical structure of man and
of the higher apes, but all the same he maintains, in conformity with his
view that all properties are evolved by exercise, that both the physical and
the intellectual superiority of man has been achieved through his having in
the course of ages exercised his faculties to an ever-increasing perfection,
while, on the other hand, the higher apes can also be trained to attain a high
standard of intelligence and a finer character. But there is still a vast difference
between Lamarck's ideas of human development and La Mettrie's and his
contemporaries' enthusiasm over the intellectual similarity between primi-
tive man and the higher animals, Lamarck maintaining that it has been given
to but few men in the whole course of the ages to achieve real intelligence,
whereas the majority have remained in a state of bestial ignorance; they
have prayed to beasts and perpetrated acts of the wildest folly; even where
a nation has attained the highest culture, this has been due to the work of a
few highly gifted persons, while the majority of their countrymen have in-
dulged in the maddest aberrations. Lamarck is without doubt thinking here
of the Reign of Terror during the Revolution, which he had witnessed at close
quarters. It was probably these memories of human degradation that deprived
him of his taste for inquiry into the characteristics of primitive man and the
MODERN BIOLOGY 315
man-apes. Possibly, too, fears of censure on the part of the Napoleonic
Government had their influence in this respect.
After a brief discussion of the term "species" in the vegetable and animal
kingdoms in relation to the mineral kingdom, wherein he states that the
mineral differs from plants and animals in not possessing individuality, and
further emphasizes the influence of environment upon the development of
species, adducing such examples as the deep-water and shore form of Ranun-
culus aquaticus, Lamarck proceeds to record his views on the nervous system
and its functions. As he deals with the same subject more fully in his sub-
sequent works, we may postpone our account of his views on these questions
and here close our resume of his Kechercbes — the work that displays his genius
and his limitations more clearly, perhaps, than any other.
In his Philosophic xpologique Lamarck discusses once more his theory of
the development of life in nature. By way of introduction he examines the
question of how much is human invention and how much is nature's own
law in natural science, and he comes to the conclusion, with Buffon, that all
systematic classifications are arbitrary products of human thought; in nature
there are only individuals, which can certainly be placed in groups in respect
of certain characteristics, but the lines between which are always arbitrarily
drawn. As regards the problem of evolution itself, he adopts the same plan
as in Kechercbes; he first describes the "degradation" throughout the animal
kingdom, and then expounds the theory as to how the organs, and therewith
the animal forms themselves, have developed by habit and way of living.
He further insists upon the importance of the essential organs for purposes
of development in contrast to the non-essential, citing the old instances of
how organs develop, to which reference has been made above, and a number
of new ones besides — sometimes quite absurd ideas, such as that the males
of the Ruminantia have acquired horns through the blood having gone to
their heads in the mating-season. In regard to the general conditions under
which life has developed, he holds that the earth has evolved continuously
and not as a result of catastrophes, as Buffon, and after him Cuvier, main-
tained, and also that no animal species have died out, except those that man
himself has eradicated, but that the fossil species that are not found at the
present time have been transformed into now existing forms. With increased
emphasis and with his criticism directed especially against Cuvier, he seeks
to prove that all animal classes are derived from one another and should
therefore be arranged in a line and not parallel or "reticularly"; nevertheless,
it is permitted for the genera in each class to form ramifications from a
common primal form. Further, in a supplement to the work he extends this
ramification theory to the classes in the Vertebrata, in that the birds are
derived from the tortoises and the mammals from the crocodiles.
32.6 THE HISTORY OF BIOLOGY
Reform of the animal system
Then follows a review of the animal system, which is one of the most
brilliant features in the whole of Lamarck's work. Here he draws up the
invertebrate system that, except for one or two alterations, has held good
ever since. He distinguishes the Infusoria from the Polypi, and the Cirri-
pedia from the MoUusca, and thus gets ten invertebrate classes: Infusoria,
Polypi, Radiata, Vermes, Insecta, Arachnida, Crustacea, Annelida, Cirri-
pedia, Mollusca. Of these classes the Radiata are now divided into two:
Ccelenterata and Echinodermata; the Polypi have been grouped with the
Coelenterata and the Cirripedia with the crayfish. A number of fresh divisions
have certainly been made in modern times, but at any rate Lamarck created
a system that, in comparison with Linnasus's invertebrate grouping, repre-
sents an extraordinary advance; and it is all the more to Lamarck's honour
that he so generously acknowledges his predecessor, whom he calls one of
the greatest scientists that have ever existed. But Lamarck was not only a
natural philosopher, he was also an expert on form, and as such he was bound
to realize the value of the preliminary work carried out by Linnaeus, although
he did not accept his hard and fast rules governing species. In this connexion
Lamarck describes the difficulties with which the systematist is overwhelmed
as a result of the aggravated chaos in scientific nomenclature; to cure this
evil he recommends that the nomenclature be fixed by international agree-
ment, and this has actually been done, though not until quite recently.
After reviewing the system of the vertebrate animals, which he has bor-
rowed from another zoologist, Dumeril, and is therefore not to be compared
in point of interest with the invertebrate system, Lamarck once more takes
up the question of the origin of man. He says that the centre of gravity in a
man standing erect is situated far in advance of the vertebras, so that muscu-
lar effort is required to hold himself upright, which indicates an origin from
quadruped animals. He drafts a theory as to man's descent from the anthro-
poid apes, but adds that this might have been so if man had not a different
origin from the animals. He has evidently not dared to draw the obvious
conclusion from his theory, but has taken refuge behind a reservation, simi-
lar to that made by Descartes in his hypothetical views on the creation.
Lamarck apparently feared that Napoleon would not have felt flattered by a
genealogy based on the orang-utan.
Theoretical speculation on life
More than half the Philosophie loologique, however, is taken up with purely
theoretical speculations on life and its manifestations, and in this sphere
Lamarck again shows his weaker side almost as much as he does in physics
and chemistry. Although in the foregoing he constantly makes nature ap-
pear as a creative power, he defines it, in his introduction to the speculative
section of the work, in the following manner: "Nature — that word that
MODERN BIOLOGY 317
is so often pronounced as if it referred to a particular being — - should not
appear to us as anything else than the comprehension of things, embracing:
(i) all physical bodies that exist, (x) the general and particular laws which
direct the changes in the condition and position of these bodies, and (3) the
motion that is current in different forms among them, eternally maintained
and renewed, infinitely varying in the products it creates . . . . " But he
is so little capable of adhering to this view that only a few pages further
on he is able to say: "Every step which Nature takes when making her
direct creations consists in organizing into cellular tissue the minute masses
of viscous or mucous substances that she finds at her disposal under favour-
able circumstances."^ A personal god could not have acted more personally.
And Lamarck's belief in creative nature is as dogmatic as was Linn^eus's
belief in God. He develops afresh his old statement that life is nothing but
motion, and that motion is produced by this wonderful and ubiquitous
ethereal fire, which is to Lamarck what the soul was to Stahl; we can apply
to the one as to the other the saying of Bonnet, that it performs anything
that one requires of it and its non-existence can never be proved. On this
basis Lamarck creates an extremely curious psychological theory. To his
mind the soul-life is a purely mechanical process, which is dependent for
its nature upon those organs that the animal in question possesses; animals
that lack muscles and nerves have practically no sense-impressions; they are
"apathetic, " they move only as a result of influences from outside, through
the ethereal fire's penetrating them and stimulating them. Animals having
a nervous system certainly receive sensible impressions, but they react to
them purely schematically and are incapable of combining the impressions
as a guide for their actions; animals that possess a brain can retain the sense-
impressions they receive and combine them to form ideas as a guide for their
actions. Lamarck's way of explaining all the manifestations of the human
soul-life — sense-impressions, ideas, and moral conceptions — with the aid
of the ubiquitous and universally applicable ethereal-electrical fluid, is in
itself of but little interest; in this he associates himself with thinkers of
the eighteenth century: Locke, Condillac, and in particular the physician
Cabanis (1757-1808), all of whom taught that ideas are exclusively based on
sense-impressions; the last-named, the most pronounced materialist of them
all, was, however, a far more trained, and therefore also a more cautious,
thinker than Lamarck, who blindly relied on his fluid, by means of which
he explained everything, while his predecessors were content to analyse cer-
tain definite phenomena in the soul-life.
Lamarck managed to complete one or two further important works in
his old age : the above-mentioned lengthy systematic survey of the Invertebrata
* Philosophic zoologiqu(, cd. cit., Part I, pp. 349, 362..
3X8 THE HISTORY OF BIOLOGY
and a work on fossil Mollusca, which is worthy to be associated with
Cuvier's contemporary works on extinct vertebrate animals. In the intro-
duction to his former work he deals for the last time with his theory of
evolution. Nature is here represented with greater emphasis than in his ear-
lier works as a creative force. "Nature is . . . an intermediary between God
and the various parts of the physical universe for the fulfilling of the divine
will." And "Nature has given to animal life the power of progressively con-
summating the organization and of developing and gradually perfecting it."
It is thus an inner striving after perfection that, besides the influence of en-
vironment, has here been the cause of evolution. This striving after evolu-
tion, which is also hinted at in his earlier writings, became, as we shall see,
a stumbling-block for Darwin, which evoked his opposition to Lamarck's
theory. It now remains to examine the hypotheses on which this wealth of
scientific production rests and the influence it had.
Influence of Buff on and Bonnet on Lamarck
The scientist by whom Lamarck as well as other biologists in France at
that period was undoubtedly most influenced was Bufi"on. We recognize this
influence in Lamarck's emphatic assertion that only individuals exist in
reality, while the categories of the classification system are products of the
mind, as also in the whole of his general conception of life as one vast evo-
lutionary process of a purely physical character;^ even the very idea of
evolution as a result of habits of life and environment we find developed
in Buffon, who cites in proof thereof the featherless face of the rook and the
padded feet of the camel. If we compare these two scientists, we find that
BufFon is without doubt superior as a thinker; he realizes the difference be-
tween hypothesis and fact, as he is aware of the limitations of natural science
— things for which Lamarck has absolutely no mind. On the other hand,
Lamarck is decidedly superior in his knowledge of form and has a far keener
eye for classification, which is certainly not exclusively due to the fact that
he was acquainted with a greater number of forms than his predecessor.
But Lamarck also learnt a good deal from Bonnet, as indeed he ex-
pressly acknowledges. The classification of the animal kingdom in one single
series was adopted by him from this source, as also the actual French expres-
sion for it — " khelle ' ' (scale). The idea of animals' ' ' degeneration ' ' through
the loss of certain organs is, however, reminiscent of Vicq d'Azyr. And,
^ Among Lamarck's precursors it is also customary to mention BENoix de Maillet (1656-
1738), for a long time French consul in Egypt and the author of a work on natural philosophy
published under the name of Telliamed (the anagram of his surname), wherein is described in
an extremely fantastic manner how the entire earth was once covered by the sea, and the ances-
tors of all existent land-animals were aquatic animals, which gradually became accustomed to
living on land. It is, however, difficult to determine what influence this work, which was treated
with contempt by Voltaire and was speedily forgotten, may have had upon Lamarck.
MODERN BIOLOGY 32.9
finally, Cuvier, with whom he was so often in controversy, has undoubtedly
exercised great influence upon him through his earlier writings; we have
already seen how Lamarck gained from them important ideas in regard to
classification, but in connexion also with problems of evolutionary history
he has undoubtedly felt the influence of his younger rival — a point that
will be more closely dealt with in connexion with the latter's own work.
But though we can thus trace outside influences in Lamarck's speculations,
it is nevertheless to his lasting credit that he formulated and elaborated as
he did the idea of origin. In that respect he is truly a pioneer of modern
biology. It is true that the theory of the heredity of qualities acquired
through the influence of environment has not stood its ground in face of
modern exact research in this field, and also that the actual method of work-
ing out the idea leaves very much to be desired. Thus, for instance, one would
have expected that when he so definitely differentiates between essential
and non-essential organs and qualities, he would have tried, following his
own method, to trace the development of the essential organs (heart, lungs,
backbone), instead of dilating upon such details as the legs and feet of waders
and swimming birds. But the idea itself is nevertheless conceived and elab-
orated not only with splendid consistency, but also with a keen eye for pe-
culiarities in the interrelation of living forms, which left all his predecessors
far behind. And indeed details can be instanced which are highly original,
as, for example, the theory of the origin of the digestive canal through in-
vagination. It is in all probability here that Haeckel, the originator of the
"gastr^a" theory and an enthusiastic admirer of Lamarck, obtained the
idea for his hypothesis, which has proved so valuable to modern embryology.
The superficiality of Lamarck's psychological speculation has already been
pointed out, but this quality he shared with many of his predecessors and
contemporaries; he is here a child of the era of enlightenment; the Supreme
Originator who was at one time creative, but afterwards inactive, and
also the subsequently omnipotent nature, are ideas that, since the days of
Voltaire, often recurred in the works of scientists of that epoch, while even
the mechanical soul-theory constantly occurs, better or worse expounded in
the writings of that period. In the purely systematical sphere, on the other
hand, Lamarck is, as previously pointed out, one of the foremost of all time,
and perhaps he has made in this sphere his most lasting, although not his
most brilliant, contribution to the development of biology.
Importance of Lamarck's life-work
We have already suggested that Lamarck's reputation in his own age was
based entirely on his work as a systematist. That he did not receive recogni-
tion as a natural philosopher was essentially due to the fact that his materi-
alistic conception of nature, which originated during the previous century,
was already out of date when he came on the scene; the two scientists who
330 THE HISTORY OF BIOLOGY
entirely controlled the development of biology at that time, and who will
be discussed in the next chapters — namely, Cuvier and Bichat — both em-
braced an entirely different theory from Lamarck's as to the intrinsic nature
of life, nor, indeed, did the latter's speculation really possess that force and
consistency which would have enabled it to hold its own against more mod-
ern directions of thought. Neither, on the other hand, did contemporary
natural philosophers pay any attention to this thought-system, which, even
from its own point of view, became rapidly out of date and, besides, was
not particularly complete in form; to one such as Schelling and his school
it must indeed have been an abomination, if only on account of its connexion
with the hated seventeenth-century materialism, and even Goethe had much
the same motive for leaving it at its worth. The day arrived, however, when
the mechanical conception of life again came into its own and when La-
marck's Philosophic :^oologique underwent a brilliant revival. The account of
this revival is reserved for a future chapter; but this much may be pointed
out here, that Lamarck's greatest admirer in modern times, Haeckel,not only
adopted his theory of evolution, but also a considerable amount of both
good and bad out of his materialistic psychology — this, too, having thus
exercised some influence up to our own day.
But though Lamarck's ideas were thus to have a future, the biological
research of his own age was being directed by a man with an entirely differ-
ent conception of nature and its phenomena, a man who possessed to a rare
degree a conception of those problems which at that time most urgently
required solution, and who was, moreover, capable of dealing with these
problems in a manner that redounded to the lasting benefit of science. This
scientist was Cuvier, one of the foremost of those who laid the foundations
of biology in the modern sense of the word.
CHAPTER II
CU VIER
GEORGES Leopold Chretien Frederic Dagobert Cuvier was born in
1769 at Montbeliard, a small town not far from Basel, which, al-
«• though entirely French, belonged at that time to the Duchy of
Wiirttemberg. He came from a French Huguenot family, which had at one
time sought refuge from religious persecution at home; his father, however,
had been an officer in French service, but in his old age had returned to his
native town, where he married and lived on a small pension given him by
the French Government. At an early age young Georges displayed brilliant
intellectual gifts; he passed through the local school with honours and dur-
ing his time there became acquainted with BufFon's writings, which he
diligently studied. The poverty of his family, however, threatened to pre-
vent him from continuing his education, when a chance opportunity procured
him free entry into the Karlsschule at Stuttgart. This one-time famous edu-
cational establishment was originally a military academy, but had been ex-
tended by the reigning Duke Karl into a college providing for the training
of Civil Service officials as well. The school was renowned for its excellent
staff of teachers and at the same time feared for the severe military discipline
exercised there under the personal supervision of the despotic Prince. Schil-
ler, the German poet of liberty, had been one of its first pupils, but had
escaped from the insufferable constraint by flight, and others had followed
his example. Cuvier, on the other hand, who was not only naturally gifted,
but also possessed a sense of discipline, got on well there; although upon
first entering the academy he had no knowledge of German, he soon became
one of the best pupils in the class for the science of State finances, which
he entered because natural science was most widely taught there for the bene-
fit of aspiring argicultural and forestry employees. The teacher of biology
here was Karl Friedrich Kielmayer (1765-1844), one of the most extraor-
dinary of German biologists, afterwards professor at Tubingen, a man who
allowed none of the courses of lectures that he gave during a long life to
be printed, though they were highly thought of, copies of them being made
and eagerly studied. He appears to have been a speculative natural scientist,
who had been influenced by Herder's ideas of a common primal type for all
living creatures and their several organs, and who consequently strongly
recommended the study of comparative anatomy. Cuvier received a thorough
331
332. THE HISTORY OF BIOLOGY
grounding at his hands and gained from him many valuable ideas,
which indeed he gratefully acknowledged throughout his life. Having suc-
cessfully passed out of the school at the age of eighteen he returned home;
he could not afford to work his way up as an unsalaried official in the Civil
Service, so he had to accept the post of tutor in a Protestant family in Nor-
mandy. Here on the Channel coast he found an entirely new animal world,
which he at once began to study with keen interest; in his spare time he
dissected all the fishes he came across and compared their structure, and with
even greater enthusiasm took up the study of the innumerable lower animal
forms that the ebb tide left stranded on the shore — molluscs, worms, and
starfish. In Linnasus's Systema Nature, which was the examination text-book
of the time, these creatures were not thoroughly dealt with; even Aristotle
had at one time displayed greater interest in marine animals, and in his writ-
ings Cuvier found not only records of their life, but also ideas suggesting
ways of comparing their different structure. He drew everything that he
studied, for he had learnt to be a clever draughtsman. Some of these pictures,
which were submitted through an aquaintance to Geoffrey Saint-Hilaire,
then newly-elected professor in Paris, proved of momentous importance for
Cuvier's future. He was summoned to Paris and within a short time was
appointed professor of comparative anatomy, although he had never dis-
sected a human body — an appointment similar to that of GeofFroy and La-
marck the year before. Thus his fortune was made and new promotions and
honours followed in rapid succession, more than space allows us to enumer-
ate. Cuvier stood especially high in Napoleon's favour; contemporary with
the Emperor in regard both to the year of his birth and to the period when
he first became eminent, he possessed something of the latter's genius for
organization; his energy was inexhaustible, he could discharge many duties
at the same time without neglecting a single detail, he was full of ideas
touching problems of organization, and he also possessed a theoretical
knowledge of statecraft which he had acquired during his school period
at Stuttgart. Thus he became " insfecteur generaV in the department of edu-
cation and carried out his duties in that post, at the same time attending to
his professorship and his science, so successfully that under his leadership
the educational system in France was thoroughly reformed and a number
of new universities founded, both in France and in its extensive subject
countries, Italy and Holland. When Napoleon fell, Cuvier became an indis-
pensable authority in the spheres of science and education; in spite of the
Catholic reaction that succeeded the Bourbon's regime, he, a Protestant, was
allowed to retain his appointments and received still further promotions,
becoming a baron and minister for Protestant ecclesiastical affairs. Through-
out this period he was wise enough to maintain his political independence,
and after the July revolution he rose still higher, becoming a peer of France.
MODERN BIOLOGY 333
By that time, however, his days were numbered; he died of cholera during
the first epidemic that ravaged Europe, in i83z. His wife survived him, but
all his children had died before him.
As a personality Cuvier has been very differently judged, both by his
contemporaries and by subsequent generations. It may be taken for granted
that one who served Napoleon with such great success was himself some-
thing of a despot, and he certainly did not escape the personal hatred that
is always the lot of such men. Bitter accusations have been made against
him even in modern times, but their truth is contradicted by the reputation
he enjoyed amongst his contemporaries. Better evidence of his true character
is provided by the unfailing dignity with which he carried on his contro-
versy against Geoffroy Saint-Hilaire, as well as by his widely attested kind-
liness and helpfulness towards younger scientists. In his political views he
was conservative, though not one of the servile type; since the appearance
of the origin-of-species theory he has been accused even of scientific conserv-
atism on account of his having maintained the immutability of species;
in this respect he must naturally be judged according to the standards of
his time and, viewed from this standpoint, his opposition to Lamarck's
theories is easily explained. As to his vital importance for the development
of biology, however, there can be no two opinions; a survey of his most
important work will confirm this.
Cuvier s co:nparative anatojny
When Cuvier set out to deal with comparative anatomy on scientific and
educational lines, he started from a point directly opposed to his predeces-
sors' line of advance. All of these had been medical men: Daubenton and
Vicq d'Azyr as well as Camper and Blumenbach; to them man was the pri-
mary object, with which all other living creatures were compared. Cuvier,
however, had begun by studying marine animals: fishes, molluscs, and
worms. Upon coming to Paris he carried out a num.ber of valuable investi-
gations, in the style of Camper, on special subjects, such as the orang-utan,
the rhinoceros, and the lemur, and later on, the Vertebrata became his chief
object of investigation. He believed his mission in life to be the creation of
a general comparative anatomy; he worked for it throughout his life and
in his other writings often referred to the forthcoming work, but it was
never completed. In preparation for it he published his lectures on compara-
tive anatomy, written down by his pupil Dumeril. The system of thought
that he elaborated in these lectures was adopted in several treatises on spe-
cial subjects: fishes, molluscs, and fossil vertebrates. Finally he published a
systematic work, Kegm animal^ based on the same principle. As a result of
these works he became, as W. Leche says, "the founder of modern compara-
tive zoology. He became so not through bringing to light a large number
of fresh facts, but rather through having introduced a new method." Much
334 THE HISTORY OF BIOLOGY
the same judgment has been expressed concerning another of the great pio-
neers whom France has given to scientific research — Lavoisier.^ This abil-
ity to create new scientific values by way of method seems to be inherent
in the French nation, with its keen and critical faculty.
The new method that Cuvier thus introduced was comparative anatomy
in the modern sense of the term. True, as he himself admits, he was not with-
out precursors — the two collaborators BufFon and Daubenton, and also
Camper, Vicq d'Azyr, and Blumenbach, had made weighty contributions
to the branch of science in question. But Cuvier's great contribution is his
consistent and far-reaching application of the comparative method, whereby
he actually created an entirely new view of the connexion of causes in nature,
in respect of both the construction of the separate individual and the mutual
relation of the various animal forms. In this sphere he has perhaps learnt
most from Aristotle, whom he resembled in his power of discovering and
comparing formal qualities of fundamental importance for the conception
of life in nature.
In his first more important work, the above-mentioned Lecons sur V anato-
mic corn-park, which came out in the years 1 799-1 805, Cuvier still to a certain
extent holds to the old point of view, the influence of his predecessors, chiefly
Daubenton and Vicq d'Azyr, being clearly apparent. But what at once strikes
one on reading this work of Cuvier's youth is the clarity and soberness of
thought that dominate his whole conception, particularly in the purely theo-
retical problems. All speculation upon the innermost essence of existence
is carefully avoided; he frankly acknowledges the powerlessness of the human
capacity for thought in this sphere and the worthlessness of those systems
of thought that earlier and contemporary natural philosophy had created
in order to fill the gaps in our knowledge of nature. In this Cuvier stands
out in sharp contrast to such scientists as Buffon, Bonnet, and Lamarck, to
say nothing of the German natural philosophers of his age. This tendency
to criticism was undoubtedly innate in Cuvier; it was certainly stimulated
by the study of Kant, whom he quotes in one place, just as, on the whole,
his acquaintance, initiated in Stuttgart, with the German world of thought
contributed towards broadening his field of vision beyond what was cus-
tomary in his countrymen at that time. The consideration of the problem
of life with which the work referred to starts is thus introduced by the em-
phatic declaration that life in its innermost essence is and must remain a
riddle: "a word that the untrained mind is ready to regard as an expression
for a special principle, although actually it can never denote anything but
the summary of the phenomena that have given rise to its formation." Then
follows a description of these phenomena, which recalls that which Hum-
^ See Part II, p. 2.65.
MODERN BIOLOGY 335
boldt gives in his early work mentioned above: we observe a state that hin-
ders the ordinary physical and chemical forces in their efforts to dissolve
the body into its simple components: this is called life, and its maintenance
requires a constant renewal of chemical components; fresh components are
absorbed by the body at the same time as others already existing in it are
given off. Finally this process ceases, whereupon death ensues, accompanied
by that dissolution of the components of the body which life had prevented.
And this life can be produced only by previous life, but the problem of the
production of life is as much beyond our grasp as is that of life itself. Cuvier
therefore does not accept the principle of spontaneous generation. Next, he
gives an account of the various components of the body — their composi-
tion and function. His account shows that he had mastered contemporary
chemistry, as developed by Lavoisier and his successors; he emphasizes the
part played by oxygen in respiration, which is expressly compared with a
process of combustion; he enumerates the simple components of the body —
carbon, hydrogen, oxygen, and nitrogen — pointing out the importance of
the last-named element in the animal organism as opposed to the vegetable
organism. It is hardly necessary to lay stress on the vast difference between
this substantiated and critical exposition of life-phenomena and their basic
structure, and Lamarck's fantastic speculations upon life. The one predeces-
sor whom Cuvier most recalls in this early work of his is without doubt
Vicq d'Azyr. But in contrast to him there stands out at once Cuvier's origi-
nality, chiefly in his conception of the object of anatomy and the consequent
arrangement of the details of his exposition; whereas in Vicq d'Azyr the
function of the organs is the essential, and the basis on which the work
rests is therefore physiological, Cuvier thrusts the form of the organs into
the foreground. He holds that respiration, whose role in the renewal of sub-
stance is the same throughout the entire animal kingdom, is performed
within the separate animal classes by means of organs which are so unlike
one another that no comparison is possible between them. And this is also
the case with the organs of motion.
His correlation theory
On the other hand, in similar animals there takes place a co-operation be-
tween the organs that makes them, as far as regards their form, entirely
dependent upon one another; the correlation between the separate organs
in the same body, which Vicq d'Azyr had already described in its main fea-
tures, is studied in detail by Cuvier and to him represents the very basis of
his conception both of animals' habits of life in nature and of their system-
atic classification. He points out that a carnivorous animal, while having a
digestive canal intended to absorb this kind of food, must also possess sharp
teeth for tearing the meat, jaws adapted to these teeth, claws for clutching
its prey, power of rapid motion, and good visual organs; a beast of prey
336 THE HISTORY OF BIOLOGY
thus never has hoofs or flat molars, for they suit only herbivorous animals.
A practised naturalist should thus be able to determine from the shape of
one single, suitably selected part of the body the whole of the animal's
structure, habits, and place in the system. And in this system, therefore,
only such animals should be grouped together as fully conform to one
another, at least in the organs that are most essential to life. The creation
of a system based entirely upon such conformity in the organs henceforth
became one of the missions in life that Cuvier never let out of sight. For
the time being, however, he contented himself with a system of grouping
that differs from the old only in that the vertebrates and the invertebrates
are distinguished from one another, and also that the lowest animals are
grouped together under the name of Zoophyta — a name of which Lamarck
strongly disapproved. Otherwise Cuvier retained the variously composed
class Vermes, and he also made his anatomical comparisons cover the entire
animal kingdom all at once. In doing so, however, he is at the very outset
careful not to extend the comparisons in detail beyond what he can vouch
for — in sharp contrast to the audacity of both Vicq d'Azyr and Lamarck,
not to speak of Geoffroy Saint-Hilaire.
His studies of fossils
During the succeeding period Cuvier applied his method to the special in-
vestigations into fishes and molluscs that have been previously mentioned,
but principally to mammals. Within this class he soon found himself engaged
in a special field of research, the study of fossil forms. As is well known,
Paris is situated in the centre of a calcareous district, in which the stone
used for building-material is particularly rich in fossils. These had already
attracted Buffon, for purposes of both observation and speculation; it was
on the basis of material gathered from this and other districts that he formed
his theory of the evolution of the earth and of the creatures living on it
(Part II, p. ii4). Cuvier, however, was the first to apply himself to a system-
atic exploration of the richly fossiliferous Paris area; with the assistance
of his friend Brogniart, he organized systematic excavations, in the course
of which the location of the fossils was closely observed and the animal
remains scattered about in each place were noted as carefully as possible.
After this Cuvier began to apply his correlation theory to fossils; for every
single bone that was discovered he searched in the neighbourhood for such
bones as appeared from their structure to belong to the first one found, if
the resultant skeleton nevertheless remained incomplete, he drew his con-
clusions from the structure of the available bones as to the habits of the
animal, and from them again as to the structure of the bones that were miss-
ing; from the bone-structure it was afterwards possible to determine the con-
struction of the soft parts. The accuracy of the method was still further
ensured by the extinct animal's skeleton being regularly compared in detail
MODERN BIOLOGY 337
with the corresponding bones of closely related existing animals. Through
this method of reconstruction, which he expounded in his famous work
Recbercbes sur les ossemens jossiles (i8ii), Cuvier created the science of palae-
ontology in the modern sense. And at the same time he largely reformed
the system of zoological classification by introducing fossil animals into it;
by account's being taken of the extinct animal forms the investigations into
the problem of affinity in the modern animal world have been far more firmly
substantiated and placed on a sounder basis than had been possible before,
and, moreover, they have led to results that to the systematists of earlier
times would have been utterly inconceivable. Of Cuvier's own investigations
in this field his comparative study of the order of elephant in particular has
won high commendation; he has here shown in the most convincing way
what results his new method is capable of giving. He begins by examining
the difference between the Indian and the African elephant, which were for-
merly grouped as a single species, but which, as he proves by comparison of
their teeth and bone-structure, are two widely different species; moreover,
he has established the fact that the extinct mammoth, of which he secured
as many remains as he possibly could, is in reality more closely related to
the Indian elephant than the latter is to the African. And, finally, he com-
pares with existing elephants a number of other extinct types, which had
either been known before and described by Buffon, or else were in the form
of newly-discovered remains; among these fossils there are some from Amer-
ica that possess knobby molars, which warrants their being formed into a
new genus, Mastodon; the members of this genus must, however, be re-
garded as true elephants, for their heavy head postulates a short neck, and
this again, as well as the long legs, show that the animal must have pos-
sessed a trunk, while from the knobby molars it may be concluded that its
food was similar to that of the hippopotamus. Generally speaking, the Pach-
ydermata especially interested Cuvier; he studied their existing forms:
rhinoceros, hippopotamus, and tapir, in comparison with their extinct an-
cestors; of these he described a number of new genera, Pal^eotherium, Dino-
therium, etc. Even the small Hyrax he removed, for anatomical reasons,
from the rodents, with which it had previously been associated, to the prox-
imity of the elephants — one of the most daring applications of his com-
parative-anatomical method. The first detailed descriptions of the American
giant sloth likewise originate from him. He also carried out some rather
sporadic studies of extinct birds and reptiles, which are of considerable value.
Cuvier as a geologist
These investigations into the existence and relationship of extinct animal
forms, however, brought Cuvier, as they had formerly brought Buffon, face
to face with the question: What changes have taken place in the character
of the earth's surface that have caused the dissimilarity between the animal
338 THE HISTORY OF BIOLOGY
world of the past and that of the present? He has tried to give the answer
in a survey of the process of the earth's development through the ages, which
forms the introduction to his Kecherches and which became his best-known
and his most discussed work. Herein, basing his argument on material de-
rived from the finest observations of earlier times, supplemented by his own,
he seeks to prove that the changes in the character of the animal world have
been caused by great catastrophes undergone by the earth's surface in pre-
historic times. He at once takes it for granted that these changes had the
character of violent catastrophes; that they were violent he considers to be
established by the fact that stratifications which, judging from the nature
of the fossils, have demonstrably taken place in the sea, are now found on
the one hand elevated to enormous heights and on the other hand over-
thrown and inverted. That all this took place with great rapidity is obvious
to his mind, not only from the sharp lines of demarcation shown by the
various strata, but also from the fact that many of them contain such extraor-
dinarily numerous animal remains that it can only be assumed that they
died a sudden death as the result of upheavals which obliterated all life for
the time being. The assumption of such catastrophic changes on the earth's
surface also affords, in Cuvier's opinion, the best explanation as to why the
animal species of ancient times have disappeared and been succeeded by new
and entirely different forms. And as a further confirmation of this assump-
tion he adduces the fact that most nations possess legends which tell of a
mighty catastrophe, a flood that drowned all living creatures, and in which
undoubtedly the mammoth and the other great land-animals living in
Europe in earlier times perished.
His catastrophe theory
It is this universally known "catastrophe theory" that without doubt
brings out both Cuvier's strength and his weakness as a natural-scientific
thinker. He does not, however, deserve any very severe censure for the ac-
tual theory of these vast volcanic upheavals, with their resultant inunda-
tions; the geological material available for observation was still somewhat
scanty and was, moreover, as far as French research was concerned, largely
gathered from the Alps, with their greatly subverted formations, which
even to this day are difficult to interpret, and which are peculiarly likely
to induce a belief in violent upheavals. But there undoubtedly existed in
Cuvier a very pronounced tendency to pursue the theories he had once set
up to their uttermost conclusions — a tendency which may well be at-
tributed to his marked aptitude for the formal side of science. Thus, he ex-
pressly declares that each stratum has its definite fossil species, which are
characteristic of it and do not exist elsewhere; the catastrophes that took
place entirely eradicated all then existent species; never has a species sur-
vived from one period to the next, so that species found in the form of
MODERN BIOLOGY 339
fossils cannot be the same as those living now; the fossil remains of lions,
bears, elephants, belong to species other than those at present existing;
fossilized human remains do not exist, the human bones that have been
declared to be such have become mixed up with fossil finds by accident.
In this connexion he maintains, in opposition to Lamarck, that the pe-
riods of geological development have not by any means been going on for
an indefinite time, but, on the contrary, during a fairly limited space of
time, and that therefore the assumption that species change through hab-
its and environment is unwarranted. If change of species were conceivable,
it would be possible, he thinks, to come across transitions between extinct
and now existing animal forms, but there are none. The immutability of
species is to Cuvier's mind an absolute fact; he has not a trace of Linnasus's
hesitation, which he expressed in his old age, in face of the difficulty of
drawing a line of demarcation between the species; according to Cuvier's
definition, species consist of "those individuals that originate from one an-
other or from common parents and those which resemble them as much as
one another." In this definition no mention is made of the creation of the
species, which, it will be remembered, Linnasus took as his starting-point,
but which, on the whole, Cuvier does not discuss at all. The assertion that
so often occurs in literature that, in his view, life has been created anew
after each catastrophe is utterly incorrect; on the contrary, he points out
that isolated parts of the earth may have been spared on each occasion when
it was laid waste, and that living creatures have propagated their species
anew from these oases, which indeed he expressly applies to the human race.
But as a rule Cuvier is not particularly interested in what might conceivably
have happened; he adheres to what he considers to be definitely proved,
leaving hypotheses to the "metaphysician." Nor is it true, as has also been
stated of him, that he allowed religious beliefs to invade the realm of sci-
ence; he certainly embraced with conviction the tenets of the Protestant
Church, whose guardian he eventually became, but in his scientific argu-
ments these doctrines play no part whatever; as a matter of fact Lamarck
refers to the Creator far more often than Cuvier. It is true that the latter
cites the First Book of Moses in support of his flood theory, but Chaldean
and Egyptian documents are quoted at the same time and with exactly the
same authority; and to ascribe historical authenticity to popular legends
was an illusion shared at that time by most professional historians.
His Regne animal
In his work Le Regne animal, distribue apres son organisation, which was pub-
lished in 1817, Cuvier develops his ideas further. In the foreword he enters
a strong protest against those who would arrange all living creatures in one
series and declares that such a method is unforgivable. He emphatically de-
nies that mammals, which come last in the system, are the lowest, or that
340 THE HISTORY OF BIOLOGY
the last mammal in the series is more perfect than the first bird. Here Cuvier
has certainly laid his finger on one of the weakest points of the whole series-
theory, as expounded by Bonnet and Lamarck, and has undeniably fore-
stalled the conception of the relativity of the degree of evolution as held in
modern times. And as an application of this doctrine of his he presents his
famous type-grouping system. According to this, the animal kingdom is di-
vided into four main groups, Vertebrata, Mollusca, Articulata, and Radiata.
Within each of these groups there is a special "ground-plan" for the con-
struction of the life-forms — a plan that appears modified in various ways
in the different systematic categories within the type. Thus, the animals
within the same type may be compared with one another, but there is no
comparison between the ground-plans of the different types. This type theory
is Cuvier's greatest contribution in the sphere of systematization and repre-
sents the farthest advance in animal classification since Linnasus; in fact,
it represents, although in a somewhat modified form, the basis of all sub-
sequent animal classification, and it is thanks to it that modern biology has
been able to lay firmer foundations for the theory of descent than Lamarck
succeeded in doing with his uniform evolutional series. But this is certainly
due to the fact that in modern times it has been possible to compare ground-
plan and organic structure even in animals belonging to different types. Here
Cuvier was far too reluctant, as indeed he was in the application of his geo-
logical theory, to draw his conclusions from the observations on which he
based his system.
Besides the account of the type theory, the work in question also con-
tains a number of observations on general scientific problems, and here, as
everywhere, Cuvier maintains the strictly critical attitude which to him
was one of the essentials of life. He is a master in not giving utterance to
more than he can stand for, and sometimes it is only in a roundabout way
that one can guess his train of thought. Thus, he repeats his above-mentioned
principle regarding life's quality of counteracting the manifestations of chem-
ical affinity in the elements that form the body, and he adds that it would
be irrational to assume that the force which acts in that way has a chemical
nature. But he enunciates no definite vitalistic theory. With equal caution
he expresses himself in regard to fertilization; how the embryo arises we
cannot tell, we can only study its subsequent development. Similarly, the
essence of the soul-life is a mystery; materialism is an arbitrary hypothesis,
"so much the more so as philosophy cannot offer any direct proof of the
true existence of matter." Here Cuvier has undoubtedly learnt from Kant;
on the other hand, his analysis of the influence of sense-impressions upon
the brain seems rather to have been influenced by Condillac and his school.
However, the knowledge with which Cuvier applies the theories of the new
chemistry to zoology represents a remarkable advance; in this respect
MODERN BIOLOGY 34I
he has laid the foundations upon which subsequent research has built
further.
His controversy with Geoffroy Saint-Hilaire
In the writings which Cuvier produced during the last years of his life there
stands out with increasing distinctness his clear, though narrow, concep-
tion of the interrelation of animal types. This is especially conspicuous in
his controversy with Geoffroy Saint-Hilaire, which attracted much atten-
tion in his own time and has been keenly debated by later generations, up
to the present day. These two had indeed been friends from youth and had
for a long time loyally collaborated. Gradually, however, their ways parted.
Cuvier insisted more and more upon the truth of his principles regarding
the immutability of species and the incomparability of types, while Geoffroy
became more and more deeply engrossed in the study of the comparison of
organs in different animal forms and speculations inferred therefrom upon
the question of one uniform type of life. Cuvier did not like personal con-
troversy; his objections to views of which he did not approve he invariably
made without personal remarks and clothed in a sometimes rather haughty,
but always courteous, style. While, then, Geoffroy for years propounded
his fantastic comparisons between the segments of the Articulata and the
vertebra;, tortoise-shell, and mussel-shell, which have been referred to in
the foregoing, Cuvier never directly opposed these, to him, absurd ideas,
but, on the other hand, formulated with increasing distinctness his own
theories and his arguments against all that contradicted them. At last, how-
ever, came the inevitable clash, in the year 1830. Geoffroy had submitted
to the Academy of Science a paper written by two younger scientists, con-
taining a detailed comparison between ink-fish and vertebrates: the ink-fish
was regarded as a vertebrate animal reflexed in the middle, with the anal
opening pressed on to the head, possessing a diaphragm, cartilages corre-
sponding to the cranial bones, and in general most of the organs peculiar
to a vertebrate animal. The essay contained a direct attack on Cuvier,
though this passage was struck out when sent to the press; but it was read
before the Academy, and therefore called for a reply. This produced from
Cuvier a courteous but sharp criticism dealing with the whole of Geoffroy's
natural-scientific speculation; by illustrating side by side the organs of an
ink-fish and of a vertebrate animal in the reflexed position, which it had been
claimed constituted the likeness between them, he demonstrated the funda-
mental difference between the organs common to both, both in structural
detail and position, showing, moreover, that many organs existing in the
one form do not occur at all in the other. But, besides this, Cuvier rejected
the entire fundamental principle on which Geoffroy based his research, at the
same time emphasizing the latter's brilliant services as an exponent of
the comparative anatomy of vertebrates. He made special reference to the
342- THE HISTORY OF BIOLOGY
similarities between bones of different vertebrates during the embryonic stage
that GeofFroy had established, maintaining that the method employed in
Geoffrey's investigations was by no means new, but originated in Aristotle,
and that Geoffroy's talk of a uniform plan for the structure of the entire
animal kingdom was mere empty words without any real meaning and with-
out any equivalent in nature. This reply greatly offended Geoffroy, where-
upon there started one of those long-drawn-out controversies so common
in scientific history, when two persons of utterly different temperament fall
foul of one another, and when the longer it lasts, the more unprofitable it
becomes. Strikingly enough, Geoffroy at once desisted from maintaining the
comparison between ink-fish and vertebrates; instead, he transferred the
whole discussion to the sphere of the vertebrates. Similarly, he replaced
the expression '' unite de plan," to which Cuvier had objected, by the phrase
" tbeorie des analogues," but at the same time emphatically declared that this
theory was entirely new; for while the old comparative anatomy concerned
itself merely with the form and function of an organ, the new theory took
for comparison all the parts of which an organ was composed. As an in-
stance of this he cited the hyoid bone in mammals, which he found to be
composed of different parts in different animals, and also the opercular bones
in fishes. Here Geoffroy was clearly referring to what we nowadays call
homology — the likeness that exists in the evolutional history of certain
organs, which warrants comparison in a manner different from what the
mere functions of these organs would justify. But unfortunately he was too
vague in his speculations to be able to give them plausible form; in the sub-
sequent discussions before the Academy, Cuvier pointed out a great number
of errors of detail even in Geoffroy's comparisons of the hyoid bones in the
vertebrates, not to mention his idea that this bone occurred in crayfish. Fur-
ther Geoffroy had a weakness for general philosophical speculation that must
have seemed utterly absurd to his sober-minded opponent. In the introduc-
tion to a book in which he collected his contributions to the discussion,
there occurs the following passage, which, like Schelling's, must be quoted
in the original: Pour cet ordre des considerations il n est plus d' animaux divers.
Un seul fait les domine, c est comme un seul etre qui apparait. 11 est, il reside
dans V Animalite; etre ahstrait, qui est tangible par nos sens sous des figures
diver ses." Such an expression of views was quite in Goethe's style — he,
too, as is well known, took part in the dispute as a warm supporter of
Geoffroy; he considered that the latter's cause was the cause of natural phi-
losophy itself, and in this he was certainly right. For if there existed in
Geoffroy's speculations advanced ideas of the greatest value even for modern
comparative anatomy, they were nevertheless an expression for that same
romantic natural philosophy, that same striving after an ideal unity in exist-
ence, which was then prevalent in Germany and which, in fact, GeofFroy
MODERN BIOLOGY 343
persistently claimed in his favour. Cuvier, on the other hand, displaying
his somewhat narrow range of vision, claimed the object of science to be
an exact knowledge of natural phenomena. It is not worth while following
the discussion between these two any further; it developed into a repetition
of old arguments and a more and more stubborn adherence to statements
once uttered.
The results of the dispute
Nevertheless, one more point in connexion with this dispute must be noted.
Among the assertions that Cuvier made and that Geoffroy at the very out-
set quotes with disapprobation, there is one which deserves attention, not
only because it shows up Cuvier's limitations, but mainly because it em-
phasizes the contrast between the origin-of-species theories of earlier times
and that of our own day. Cuvier says of the ink-fish: "They have not resulted
from the development of other animals, nor has their own development
produced any animal higher than themselves." In face of the modern theory
of evolution the former of these statements is undoubtedly untrue, whereas
the latter is correct. To Lamarck and Geoffroy both statements were equally
untrue and they became even more excited over the latter than over the for-
mer. For what they were particularly in search of was just that connecting
link between the highest form in each class and the lowest type in the next
one — ink-fish and fishes, tortoises and birds, to name two examples. On this
rock the earlier theories of origin regularly suffered shipwreck. The fact that
the modern historian of evolution has learnt instead to search backwards in
the evolutional series, in order to find among more primitive forms primary
types for the separate, highly specialized groups, has been rendered possible
owing to modern zoology's having accepted Cuvier's type theory, which
avoids direct comparison between highly developed life-forms of different
types — but thanks also to embryology, which Geoffroy endeavoured, though
without success, to make the basis of his theory of comparison between
organs of the same kind in different animal forms. It happens then, that both
the parties to the dispute of 1830 have contributed to the creation of the
modern view of natural evolution.
CHAPTER III
BICHAT AND HIS TISSUE THEORY
IN THE IMMEDIATELY PRECEDING SECTION of this worlc One chapter (chap-
ter v) was devoted to giving an account of the two mutually opposed
ideas as to the nature of life that were prevalent during the early part of
the eighteenth century: the mechanistic, which conceived the phenomena of
life from the purely mechanical point of view, and the vitalistic, which, rep-
resented by Stahl and his pupils, saw in the soul the real entity of life and
regarded the body as existing for and through the soul. Curiously enough,
Stahl's doctrine, the most markedly vitalistic of them all, won support
particularly in France, where it was preserved and further developed by the
medical faculty at Montpellier. It was especially Stahl's idea of the complex
chemical composition of the body and the easy decomposability of its constit-
uent parts, and the peculiar structure of them characteristic of different
beings, that was developed by the Montpellier school. On the other hand,
these scientists paid less attention to Stahl's speculations on the soul itself;
rather, it was life, the life-force, that was believed to be the binding force
that prevents the chemical components of the body from disintegrating. We
have seen Stahl's theory recur in this form both in Humboldt and in Cuvier.
In actual fact the sharp distinction between mechanism and vitalism was to
a certain extent removed towards the close of the eighteenth century; the
progress of chemistry made it necessary to consider other functions in the
body besides the purely motive phenomena — ■ a fact that even the most con-
vinced mechanists eventually had to realize; while, on the other hand, a
number of active natural forces were discovered — primarily the electric and
the magnetic — of which earlier ages knew nothing and in face of which bi-
ology — whether vitalistic or mechanistic — was bound to adopt a definite
attitude. As examples of the influence of these new discoveries may be men-
tioned, on one hand, Galvani's experiments with electrical phenomena in
the organism, which were continued by Humboldt and others, and on the
other Mesmer's investigations into "animal magnetism," or what we should
nowadays call hypnotic phenomena. As a result of all these complications,
that age's conception of life-phenomena became a mere groping in the dark; it
was only after the discovery of the law of the indestructibility of energy that
biology also gained a fresh basis on which to build, as a result of which it
became possible to form a fresh mechanical conception of life-phenomena.
344
MODERN BIOLOGY 345
Among the scientists of the MontpeJlier school who disputed the pre-
vailing mechanistic theory of life, which was maintained chiefly on the
authority of Boerhaave, may be mentioned Theopiiile de Bordeu (1711-76).
The son of a doctor living in the south of France, he settled down in practice,
after taking his degree, first in his home district and later in Paris. He wrote
an account of his views on life-phenomena in a work entitled On the Glands.
Contemporary physiologists of the mechanistic school of thought believed
that glandular secretion was due simply to the mechanical pressure of adjacent
muscles. Through a series of careful experiments and extensive investigations
based thereon, Bordeu proves that mechanical compression cannot produce
glandular secretion. This is due rather to the direct influence of the nerves
leading to the glands. Through this nervous influence the supply of blood
to the gland is increased and by means of a purely mechanical arrangement —
Bordeu believes that he found openings capable of expanding or closing
through the influence of the nerves — the follicles of that gland absorb out
of the blood such fluids as are characteristic of the secretion. This individual
power of the gland to absorb fluids that are suitable to it Bordeu names
"sensation" and he ascribes to each organ in the body a special power of self-
operation, a "tact," as he calls it; the stomach absorbs certain substances,
and reacts against others by the process of vomiting; the eye has its special
reaction against the outside world, and likewise the ear. Life proceeds as
the result of co-operation between the individual action of all the organs.
The brain and the nervous system control this co-operation; their action is
expressed in the alternate contraction and expansion of their mass. Although
Bordeu evinces great admiration for Stahl, it is nevertheless with extreme
caution that he expresses any opinion on the question of the soul's relation
to the body, just as in general he avoids entering into more abstract regions
of thought.
We find, on the other hand, speculations of a markedly natural-philo-
sophical character in a somewhat later pupil of the Montpellier school, Paul
Joseph Barthez (1734-1806). Hewas first of all a practitioner, then professor
at Montpellier, and finally chancellor of its university. Being of a pugnacious
and irascible nature, he became involved in many a dispute, especially after
he had taken sides with the aristocracy during the Revolution. Having been
deprived of his post, he lived for a time as a private individual. He published
his theoretical opinions in a work bearing the striking title of Science de
Vhomme. By way of introduction hj gives an analysis of causality, which he
afterwards examines with special reference to the cause of life. Barthez finds
the ultimate cause of life to be inexplicable and considers that neither
Boerhaave' s nor Stahl 's theories are satisfactory hypotheses or of any use
to medical science; in their place he assumes a special "life-principle" as the
foundation for the vital manifestations of all living creatures. In man this
346 THE HISTORY OF BIOLOGY
principle does not coincide with the conscious psychical life; it is rather a
kind of general force, which contains within itself irritability, sensibility, and
the other vital phenomena described by physiologists of that period. This
speculative tendency, it appears, follows a line of thought exactly opposed to
that which Bordeu pursued; he tried to regard the vital manifestations of the
different organs as isolated phenomena, while Barthez sought above all for
one common principle that would hold good for all manifestations of life.
These and other members of the Montpellier school during the eighteenth
century would, however, scarcely deserve mention beyond the borders of
France had not their work formed the basis on which Bichat built further.
Marie pRANgois Xavier Bichat was born in 1771 at Jura, in eastern France.
His father was a doctor of repute, and the son was from early youth destined
for the same profession; after finishing school he studied surgery at a hospital
at Lyons, but when that city was destroyed during the wars of the Revolu-
tion, he betook himself to Paris. There he found a patron in the surgeon
Desault, with whom he worked both as a surgeon and as an anatomist. After
the death of his benefactor, in 1795, he spent a couple of years editing his
writings; in return he found in Desault 's widow a maternal friend and a
practical adviser. Bichat displayed throughout his short life an enthusiasm
for science that has hardly ever been equalled; although he lived through the
most exciting events of the French Revolution that occurred in his immediate
neighbourhood, he was nevertheless able to devote himself entirely to his
anatomical works. He spent his days and quite often his nights in the ana-
tomical theatre, in order not to waste time. Nor did he care much about
promotion; in 1797, however, he began to give lectures and four years later
was appointed to a professorial chair, without having applied for it. More-
over, he carried on very intensive work as an author and took part whole-
heartedly in the life of the medical faculty. In the spring of i8oi, however,
he was attacked by a malignant fever — whether as a result of septic poison-
ing or whether owing to some other infection is apparently not known —
and he died in spite of the utmost care of his friends and colleagues; at the
time of his death he had not yet reached his thirty-first year.
Bichat' s character is described by his contemporaries as mild, modest,
and unselfish. His writings testify to a general knowledge that was surpris-
ingly extensive for such a young and extremely busy man, and yet his is the
work by no means of an unpractical bookworm, but of one who took a deep
interest in life and also observed a great deal in his fellow men. His works
were written during the last four years of his life; in the early maturity of
his creative powers he resembled Linnreus, as also in the fact that his genius
was primarily formal and systematic. Bichat has introduced a new system
into the science of anatomy and it is in this fact that his chief greatness lies.
In his writings Bichat shows himself above all a medical man; the func-
MODERN BIOLOGY 347
tions of the body are invariably described in close relation to its morbid
changes and to the manner in which they should be treated. Pathological
anatomy engages his interest quite as much as normal anatomy, and post-
mortem examinations formed a considerable part of his practical work. He
studied the various parts of the body in both its healthy and its diseased
state, employing a number of different methods for the purpose; besides
dissection he mentions drying, cooking, and maceration, as well as treat-
ment with acids, alkalis, and alcohol. On the other hand, he did not use a
microscope, for he thought that this only gave rise to fallacies and delusions.
And yet it is as the founder of a science of microscopy that he won his highest
fame. Another peculiar fact about him was that he despised the illustration
as a means of reproducing the results of research; in his view, all represen-
tations, even plastic, illustrate only in an imperfect and misleading manner
the facts which the research-worker wishes to convey. His writings do not
contain a single illustration.
Bichat' s conception of life
Bichat's conception of life has always been regarded as vitalistic. Indeed, his
theoretical fundamental view is unquestionably reminiscent of Stahl; life,
says he, is "the sum of the functions that resist death." It is a far cry, how-
ever, between Bichat's so-called vitalism and Stahl's; the latter's theory of
the soul as the ultimate end and conservator of the body Bichat strongly
denies. Stahl, he declares, had realized the incompatibility between physical
laws and animal functions, but because the soul was everything to him in
explaining the functions of life, he failed to discover the laws of life. With
equal emphasis, however, Bichat rejects Boerhaave's theory that life should
be regarded as a purely mechanical process. "The true essence of life is un-
known; it can only be studied through the phenomena it manifests"; and
among these phenomena the most conspicuous is that previously men-
tioned — that it resists the influence of those forces which strive to disinte-
grate the body and which achieve their object as soon as life has departed.^
As is well known, Stahl laid special stress upon the complexity of the body's
chemical composition and its consequent easydecomposability as being some-
thing essential to life; this truth was appreciated by Bichat more than by
any of his predecessors and was further developed on the basis of the epoch-
making discoveries in chemistry in his own age. The primary lesson he learnt
from Stahl, however, is the importance that different structural conditions
have for the functions of the organism; in fact, the theory of structure repre-
sents Bichat's greatest contribution to the development of biology; it forms
one of the corner-stones on which our conception of life and its manifes-
tations rests.
^ This definition recalls Humboldc's early ideas referred co above and may certainly be
derived from the same source.
348 THE HISTORY OF BIOLOGY
His classification of tissues
According to Bichat's classification, the body is built up of tissues, which
may be grouped in systems — for example, the bone system, the cartilage
system, the muscle system. An organ is composed of several systems (e.g., the
stomach, the lungs, the brain); several organs form an apparatus (e.g., the
respiratory apparatus, the digestive apparatus). The knowledge of the tissue
systems forms what Bichat calls "general anatomy," which he discusses in
an important work;- upon this knowledge should be based the theory of
organs or, as he calls it, descriptive anatomy. Bichat claims that this method
of research and investigation is new, for, as he adds with justifiable self-
appreciation, general anatomy hardly existed before he produced his works
on the subject.
The tissues, Bichat declares, are the true conservators of the life of the
body. He distinguishes between twenty-one diff^erent kinds of tissues — ■
namely, (i) cellular (closely corresponding to what is now called retiform
connective tissue); (z) the nervous tissue of animal life; (3) the nervous
tissue of organic life; (4) arterial; (5) venous; (6) the tissue of exhalation;
(7) absorbent; (8) bone-tissue; (9) medullary tissue (in the bones); (10)
cartilaginous; (11) fibrous; (li) fibrocartilaginous; (13) animal musculature;
(14) organic musculature; (15) mucous tissue; (16) serous; (17) synovial;
(18) glandular; (19) dermoid; (xo) epidermoid (dermis and epidermis);
(21) capillary tissue. These tissues, however, are by no means alike every-
where; rather, they invariably possess the power to adapt themselves to the
organs in which they are incorporated. The tissues are the true conservators
of life; not each individual organ, as Bordeu asserted, but each individual tis-
sue has individual life. Therefore diseases, in so far as they attack individual
organs, are localized in their tissues; in abdominal catarrh it is the mucous
membrane that is affected and not the m.uscles in the abdominal wall; in
inflammation in the brain it is in most cases the cerebral membrane that is
the seat of the disease. "If we would study a bodily function, we must con-
sider the organ which performs that function from a general point of view,
but if we would become acquainted with the vital qualities of the organ, we
must disintegrate it" — that is, into the tissues of which it is formed. The
tissues are thus the vehicles of life; in maintaining this view Bichat definitely
dissociates himself from a number of earlier and contemporary scientists,
who considered the fluids to be the true vital elements of the body.^ But the
vital qualities are not identical with the actual structure, for this still
remains after life has departed; not even the fluids of the body are the same
after death, and if we analyse them chemically we find only an equivalent
^ The Traiti des membranes, cited in the bibliography, may be regarded as a preliminary
study to this work.
' See, for instance, Sommerring's above-mentioned theory on the cerebral fluids.
MODERN BIOLOGY 349
to the anatomical nature of the dead body. Life consists rather in certain
qualities, which the living tissues possess and which are not found in inani-
mate nature. Here Bichat takes as his starting-point Haller's previously-
mentioned theories of irritability and sensibility and he develops them
further, expressing great appreciation of Haller, who, to his mind, had a
far more correct view of life than Stahl. According to Bichat, sensibility
is the characteristic quality of the nervous system; the muscular system dis-
plays a quality that he calls contractility; this has different characteristics
in different organs and should not be confused with the tensibility that the
tissues possess independently of life. But life manifests itself not only in these
qualities, but in still another phenomenon, unknown in inanimate nature;
this is called sympathy and expresses itself in the effect that the vital func-
tions of the various organs have upon one another in conditions of sickness
and health. Bichat made serious attempts to ascertain the nature of these
vital phenomena by experimenting with living organs under various con-
ditions. Thus he tried to analyse especially muscular contractility and dis-
tinguishes several categories thereof — namely, he holds that the muscle
comes into action: (i) as the result of impulses from the brain received
through the nerves — that is, normal contractility; it ceases if the nerve is
severed; (i) through chemical or physical influences — that is, organic and
sensible contractility or irritability; it ceases if the muscle is deadened (e.g.,
by opium); (3) through the fluids which the vascular systems convey to the
muscle and which distend its minutest parts — that is, passive contractility
or tonicity; it ceases as a result of death; (4) finally, the muscle contracts on
being severed — that is, the contractility of tissue itself, which only ceases
as a result of putrefaction. Of sensibility he distinguishes two categories —
namely, "organic," which consists in the power of receiving an impression,
and "animal," which not only receives the impression, but conveys it
farther to a common centre and is thus a higher category of the previous one.
Organic and anbnal life
The contrasted ideas, organic and animal, frequently referred to above, play
an important part in Bichat's explanation of life. "Organic" are vegetable life
and the unconscious life of animals; "animal" are the functions in animals
that are controlled by the will of the individual and are consequently the
more developed the higher the life is. Even in modern times one sometimes
differentiates between animal organs, among which are included especially
the nervous system and the motive organs, and the vegetative, among which
are included the digestive, circulatory, respiratory, and excretal organs.
Bichat, however, carries this differentiation to most absurd extremes when
he consistently speaks of "the two lives," the animal and the organic, and
assures us that the former's organs are always symmetrical and the latter's
unsymmetrical, much labour being spent on trying to prove that lungs and
350 THE HISTORY OF BIOLOGY
kidneys are in reality unpaired. Similarly, he endeavours to show that the
animal functions are always "harmonious," while the organic are "dis-
cordant," by which is meant that it makes no difference if one lung functions
more or less than the other, whereas dissimilarity in the visual or auditory
organs causes serious disturbances; the lack of a gift for music Bichat con-
siders to be due to the fact that the ears possess different powers of hearing.
He does not include the sexual organs in either category, because they serve
the genus and not the individual. Of the psychical qualities, the intelligence
belongs to animal life, while the passions are derived from organic life, from
disturbances in the digestion and the blood-circulation. The community is
thus only a development of animal life, while the passions have brought about
all human disasters — revolutions and reigns of terror. In all this Bichat
shows an inclination for sophistry, which not infrequently accompanies a
highly developed genius for the purely formal. Several others of his sys-
tematic divisions are also by no means wholly successful. At all events, if
only for the new system that he introduced into anatomical science, Bichat
must be counted as one of the greatest pioneers of that science that have ever
lived. Considerations of space forbid a more detailed account of his thorough
exposition of the different tissue systems which he gives in his general anat-
omy, as also of his application of it to the theory of organs in his descrip-
tive anatomy. His work contains no histology in the modern sense, but this
is only natural, as he refuses to learn from microscopical observation; on
the contrary, he dismisses with some compassion Leeuwenhoek's attempts at
determining the form and size of muscular fibrillar; the true nature of muscu-
lar fibre is unknown, and that is all there is to be said about it. He is far
more interested in the chemical composition of the tissues, as far as it was
possible to ascertain it in those days, and in their condition under processes
of drying and maceration. It is at any rate the topography of the tissues that
chiefly engages his attention; their finer quality did not concern him; for
instance, Malpighi certainly knew more about the structure of the brain
than he did. Bichat 's greatness, then, lies in his having so convincingly
proved the quality of the tissues as fundamental constituents of the body and
its functions. He thereby placed the study of the phenomena of life on a defi-
nite basis, the value of which is best realized if we compare his tissue theory
with the fantastic ideas of a "nervous fluid" and "microscopical life-units,"
in which the works of even the most brilliant biologists of the imme-
diately preceding epoch abound. Even the terms "sensibility" and "con-
tractility," which were invented by him, have been incorporated in modern
terminology. And although his ideas of the application of physics and chem-
istry to biology must appear primitive to a modern reader, still, he had the
eye of a genius for essentials in the contrast between animate and inanimate
matter, which many a modern biologist has lacked. That he so strongly
MODERN BIOLOGY 35I
maintained this difference was certainly well justified as a reaction against
those clumsy mechanical theories of life which were then being propounded
by Lamarck and others. Nor indeed can it be said that Bichat prostituted his
contractility idea to metaphysical or mystical speculations. His mind was,
in fact, trained through studying the sceptical philosophers of the eighteenth
century — he quotes both Condillac and Cabanis — and their criticism
formed a sound counterbalance to the bold ideas which he learnt from Stahl
and his school. Bichat knew to a fair degree how to retain the best of what
he learnt from his predecessors and how to establish on the basis of the
knowledge thus gained a consummate field of research of his own.
CHAPTER IV
cuvier's younger contemporaries
THE FIRST HALF of the eighteenth century shows in general a lively de-
velopment in the sphere of biology. The splendid progress made
simultaneously in physics and chemistry created innumerable fresh
problems also in the biological sciences; voyages of geographical explora-
tion, which were made to hitherto unknown lands on the precedent of
Humboldt, resulted in new material for investigation, which broadened
existing ideas and broke down old systematical barriers — ■ we need only
mention such animals as the duck-billed platypus and the lung-fish in order
to show clearly the importance of these discoveries — • and, finally, the vast
technical and economic progress of that epoch awakened an interest in the
study of nature, which also proved of benefit to biology. Among the techni-
cal inventions that belong to this period may first of all be mentioned the
improvement in the construction of the microscope, which alone has given
mankind a knowledge of a whole series of hitherto unknown life-forms; the
economic progress, again, rendered possible the instituting of collections
such as earlier times had never dreamt of, as well as the carrying out of costly
experiments on a large scale. As a result of all these circumstances, of which
many keen scientists took full advantage, biology achieved more and more
brilliant results as the years went by, with the consequent enhancement of
its general cultural reputation — in spite of the indignant protests and the
scornful rejection of the idealistic philosophers.
Progress of comparative anatomy
One branch of biological research which, more than any other, made rapid
strides during the period under discussion was that of descriptive and com-
parative anatomy. Cuvier, its most brilliant precursor, had many pupils in
different countries, both direct and indirect, who carried on the numerous
ideas he brought out, and besides these there were many others whose re-
search work resulted in valuable contributions towards the progress of
science. A number of the most representative of these scientists will be dealt
with in the present chapter.
Carl Asmund Rudolphi was born in 1771 in Stockholm, of German
parents. He studied medicine at Greifswald, at the German university of
Swedish Pomerania, becoming professor in anatomy first there and afterwards
in Berlin, where he worked until his death, in 1831. He founded the Berlin
352-
MODERN BIOLOGY 353
zoological museum, now one of the finest in the world, and did much ex-
tremely successful educational work; he was highly esteemed by pupils and
friends on account of his zeal for science and his noble, almost supersensitive
temperament. He could never get himself to perform vivisections and once
declared that even the prospect of world-wide fame would not induce him to
possess the insensitiveness of a Brunner.' At the same time he was severely
critical towards others as well as towards himself and laboured hard for
the abolition of the mysticism that natural philosophy had introduced into
biology, so that his writings, in spite of a number of inaccuracies, give on
the whole an impression of solid reality and of being far more modern than
those of many of his famous contemporaries.
Ktidolphf s work on -parasites
It is in three particular branches of biology that Rudolphi has made valuable
contributions: parasitic research, comparative anatomy, and physiology. In
the first-named he is a pioneer; his work Entozporum historia naturalis has so
considerably widened the knowledge of intestinal worms which Pallas
founded that all subsequent research has been based on it; this work is the
result of investigations into numerous germ-carrying animals and gives de-
tailed accounts of the appearance and conditions of life of the parasites
existing in them. Through this work the number of known species of intesti-
nal parasites has tripled. But while Pallas believed that the parasites or their
eggs enter the host from outside, Rudolphi is convinced that they are pro-
duced by diseased processes inside the hosts — a false idea, which is all the
more curious because otherwise he most emphatically denies the possibility
of spontaneous generation. In these circumstances it is natural that his
account of the evolutional history of the intestinal parasites should be the
weakest part of his work and far inferior to the masterly description that
he gives of the different forms.
In a collection of short essays Rudolphi has recorded a number of valu-
able investigations in comparative anatomy. Of special interest are his com-
parative microscopical studies of the intestinal villi in different vertebrates.
He gives an account in this work of a large number of different forms of these
appendices, thereby increasing not only the knowledge of the tissue theory
as created by Bichat, but also the use of the microscope. This investigation
therefore deserves to be remembered as one of the first in the sphere of com-
parative histology. Another useful work of his was the study of the cerebral
cavities, wherein he attacked Sommerring's above-mentioned theory of the
cerebral fluid's being the intellectual organ and his belief in connexion there-
with that the nerve-fibres end in these cavities. Rudolphi considers that the
' JoHAN Conrad Brunner (165 3-1737), whose name is preserved in the glands of the
duodenum called after him, made some well-known experiments with the extirpation of the
pancreas of live dogs.
354 THE HISTORY OF BIOLOGY
entire brain is the "intellectual organ" and maintains, in contrast to the
attempts of earlier times to localize the soul, that such a complex phenome-
non may very well require a complex organ as its foundation. On the other
hand, Rudolphi's attempt at a new systematic grouping of the animal king-
dom, into nerveless, single-nerved, and double-nerved, was not very success-
ful and was forgotten long ago — a scheme, in fact, which had already been
rejected by his contemporaries and which had to give way to Cuvier's better
and more natural basis of classification.
His text-book on physiology
Rudolphi's most important work, however, was his Grundriss der Physiologic,
which occupied his old age and was still unfinished at his death. This work
best displays his great knowledge, founded on his own experiences and his
wide reading, as well as his critical faculty and elevated mode of thought.
Physiology, he says, is "the doctrine of the human organism." An organism
without life is unthinkable; when the one is created, the second must be
present; a dead body is not an organism, but only the remains of one. The
classifying science of physiology is therefore anthropology: here Rudolphi,
like Blumenbach, strongly insists upon man's dissimilarity from the apes,
but considers in opposition to the latter that the human genus should be
divided into species, not races. In this connexion he declares that human
beings cannot have originated from one single pair — a sentiment which,
during that reactionary period, it certainly required some courage to express.
The chapter on anatomy that follows this section is one of the most brilliant
parts of the work; the clear and concise manner in which he expounds the
composition of the human body was unrivalled at the time; as compared
with Bichat's tissue theory it represents a great advance, on account of both
its simple and concise grouping of the tissues and its sound criticism; here we
find no fantastic theories of life, no absurd speculations on symmetry, but a
clear and sober account of the different parts of the body, which is mostly
consistent with modern conceptions. In regard to the essence of life, Rudolphi
associates himself most closely with Reil's theory of life as bound up with
the form and mixture of matter, while Oken's and Schelling's extravagant
ideas are utterly repudiated. Likewise, he rejects Stahl's theory of the soul
as a cause of bodily phenomena: "Das Dasein oder das Hinxutreten eines Geistes
oder einer Seele xum Korper erkldrt uns das Leben nicht im geringsten. ' ' On the other
hand, he strongly emphasizes the importance of the chemical processes for
the vital functions, in this associating himself with Berzelius's animal
chemistry, which, thanks to his childhood's having been spent in Sweden,
Rudolphi was able to read in the original. His account of the functions of
the nervous system and the sensory organs is an extremely careful piece of
work, which sharply criticizes all the mystical nonsense that was prevalent
at that time: animal magnetism, interpretation of dreams, divining-rods,
MODERN BIOLOGY 355
and the like. Occasionally, however, his criticism goes beyond the mark; he
expresses doubt not only of Gall's theory of the nerves' leading to the grey
matter of the brain, but also of the existence of sensory and motor nerves.
The caution with which he treats current theories is at any rate attractive.
Space forbids a more detailed account of his exposition of the digestive
organs, respiration, and the musculature; these organs and their functions are
described with the same thoroughness and care that marked his previous
chapter. The whole work testifies, even in its incomplete form, to the strides
that exact research had already made during the first decades of the nineteenth
century, as regards both methods and results, foreshadowing the immense
successes of subsequent periods.
Contemporary with Rudolphi there was working in Germany a scientist
who made important contributions in the sphere of exact biology, although
in his theoretical conceptions he maintained the point of view of the natural
philosophy of the time. Johann Friedrich Meckel was born at Halle in
1781. Both his father and his grandfather had been professors in anatomy
there; both had by their keenness and insight improved the anatomical
collection existing there — especially the father, who had even declared in
his will that his skeleton was to be mounted and set up in the museum.
Young Meckel followed in their footsteps; having matriculated and taken
his degree at Halle, he worked for a year or two with Cuvier and afterwards
made a tour through Europe. At the age of twenty-five he was appointed
professor in his native town, where he worked until his death, in 1833.
During his lifetime he exercised great influence both as a teacher and as a re-
search-worker, and not least as a result of his having undertaken the publica-
tion of Reil's above-mentioned Archiv, in which he developed his own ideas
and gave accounts of a number of special investigations. These partly fall
within the sphere of descriptive and comparative anatomy, and partly they arc
purely speculative and philosophical. In the former sphere Meckel proved a
worthy pupil of Cuvier, while in the theoretical sphere he was undoubtedly
influenced by GeofFroy Saint-Hilaire. The name of "the German Cuvier,"
which his contemporaries gave him, thus only partially corresponds to his
point of view; what made him most worthy of the title was the work he did
by exhortation and example to introduce into Germany the study of com-
parative anatomy, which in course of time was to reach its highest develop-
ment in that very country. Amongst his contemporaries, at any rate, there
was not one, with the exception of Cuvier, who had mastered the anatomy of
all animal forms, both higher and lower, so thoroughly as he, though his most
important investigations he carried out in the field of vertebrate anatomy.
Meckel's system of comparative anatomy
Of these specialized investigations of Meckel's the most exhaustive are his
anatomical monographs on the ornithorynchus and the cassowary, but
356 THE HISTORY OF BIOLOGY
besides these he has made considerable contributions in several other spheres of
anatomy: investigations into the development of the nervous system and the
intestinal tube in the embryonic stage, into the structure of the brain of birds,
into the intestinal villi, which reasons of space make it impossible to discuss
in detail. Finally, in his later years he collected the results of his researches
in a large work entitled System der vergleichenden Anatomie, which, like Rudol-
phi's Physiology, was never finished. The first part of it forms a summary of
Meckel's theoretical speculations; the following sections deal with the
structure of the individual organs.
If we turn from reading Rudolphi's Physiology to Meckel's general
anatomy, the first impression will undoubtedly be that we have taken a
long step backwards in time — from a critical and almost modern method
of presentation back to romantic natural philosophy. The very foreword
contains the statement that the " Bildungsgesetze" which govern the animal
kingdom may be grouped under two main principles, "multiplicity" and
"unity," the latter also being termed "reduction." And the exposition of
these "formative laws" is introduced with the assertion that the form of the
animal may be regarded either in itself and with reference to the physical
force which is its origin, or with reference to the purpose intended to be
served thereby and the creative spiritual force that forms the basis thereof.
This at once is far more reminiscent of Schelling than of modern natural
science, and yet we constantly come across proofs of the author's many-sided
and radical knowledge of anatomy and of his genius for combining acquired
facts. By the formative law of multiplicity is meant all those qualities which
distinguish the life-forms from one another, and herein are included not only
the characters that differentiate the species, genera, and higher groups, but
also the qualities of the individual organs in the same and in different
animals and the changes in them, such as are brought about by age, habits of
life, and heredity. Under this heading, then, comes descriptive anatomy, while
the "reduction" law embraces comparative anatomy, or, as Meckel says, the
proofs that all formations in the animal kingdom are variations of one single
type — that is, the same idea that Geoffroy and Goethe tried to develop.
Law of multiplicity
Under the law of multiplicity is described, to start with, the body's for-
mation of tissues — a chapter in which Meckel does not compare well with
Rudolphi in clarity and conciseness. As the fundamental substance for all
the parts of the body he gives a solid matter, shaped like minute globes,
which lie embedded in a fluid; these are clearly visible in the lower animals
and in the embryos of higher animals, while in the latter themselves the fluid
is coagulated and in conjunction with the globes forms fibres, membranes,
and tissues. In this speculation Meckel is without doubt influenced by Caspar
Friedrich Wolff, whom he held in high esteem and whose writings he trans-
MODERN BIOLOGY 357
lated into German. As regards the system, Meckel to a certain extent adopts
an attitude of indecision; on the one hand he has to accept Cuvier's types,
but on the other the whole object is to prove the possibility of one single
primal type; consequently Meckel rejects the definite line that both Cuvier
and Lamarck draw between vetebrates and invertebrates, and as an inter-
mediary form between them he places the ink-fish, whose carapace is declared
to be the rudiment of a backbone. Further, Meckel, like Lamarck, believes
in a common spontaneous generation whereby a number of lower life-forms
arise in various parts of the world and thus increase the number of existing
species. In this, as in his general idea of the origin of life-forms, Meckel in-
clines towards Lamarck, his indebtedness to whom he acknowledges when
quoting him. Each of them sought to produce a "history of natural creation,"
and it must be admitted that on this subject Meckel was able to derive ad-
vantage both from the work of his predecessors and from his own thorough
knowledge of anatomy. Meckel's theory of origin thus contains many in-
teresting and suggestive ideas of importance for the future of science, but
it certainly contains also masses of weird fancies and ridiculous conclusions.
What distinguishes Meckel's theory from Lamarck's — and even from Dar-
win's — is the fact that he does not assume one single cause of evolution, but
a number of causes, and his exposition herein lacks the easy comprehensibil-
ity that characterizes both his predecessor's and his successor's work, which,
indeed, explains why it is that he failed to win the same degree of popu-
larity that they did. Among the causes of evolution, it is true, Meckel, like
Lamarck, lays great stress on the influence of habit and environment, or, as
he expresses it, the formative influence of mechanical forces. In this connexion
he quotes stories of how bobtailed equine and canine races have been created
as a result of the tails of the animals' ancestors having been docked, and in
the same way mechanical pressure in the course of ages has produced the
numerous interlacings and various divisions of the digestive canal, as also
other changes in the internal organs. And he ascribes similar transforming
force to light, heat, and electricity; in particular, the electrolysis of fluids,
which was then newly discovered, leads him to indulge in fantastic specu-
lations upon the effect of this force on the development of life-forms. More-
over, he drags into his theory of the formation of species the entire category
of mostly unknown phenomena that give rise to malformations; thus, he
cites the old belief that mothers can give birth to malformed children after
"getting a fright," as at least a plausible cause of the appearance of new life-
forms, and he finally mentions hybridization as an important cause of the
arising of fresh species. To this factor in the evolution of life, however,
which, as is well known, has excited special attention in modern times, he
attributes utterly irrational eff^ects: he believes in old tales of a cross between
a cat and a hare, a cock and a duck. There is indeed far better justification
358 THE HISTORY OF BIOLOGY
for his expressly ascribing the abundance of species in the insect class to
hybridization.
Law of unity
All the speculations just referred to, Meckel includes in the sphere of the
law of multiplicity. As already mentioned, under the law of reduction come
the proofs of the unity of the life-type. Here Meckel displays the whole of
his extensive knowledge of anatomy, both good and bad; here he presents
his most brilliant ideas and also indulges in the most ridiculous absurdities.
The latter necessarily result from the false preconception of one single life-
type and evolutional series, the same fatal constructions of thought, in fact,
which led Lamarck and GeofFroy to such wild delusions. And it may be said
that herein Meckel outrivals them to the same extent as his anatomical
knowledge is more extensive than theirs. It is hardly worth while going
further with him along these erratic courses; as when he compares the shell
of the tortoise and of insects, or the papillas on the tongue of cats and of
snails, or when he likens the double malformations occurring in man, now
generally called Siamese twins, to a colony of polypi. It would be better to
ponder over the numerous ideas of great value for the future to which he gives
expression in this connexion. Among these ideas that have been generally
adopted in modern times may be mentioned his view that the lungs of the
land vertebrates are derived from the air-bladders of fishes, his comparison
between the male and the female sexual organs and his derivation of their
several parts from an indifferent embryonic stage, his comparison of the
segmentation of worms and articulates with metamerism in the vertebrates,
and, above all, his foreshadowing of Haeckel's biogenetical organic law,
when he declares that each higher animal during its embryonic development
passes through the same forms as those that are lower in the evolutional
series possess when fully formed. This theory of "the development of the
special organism in accordance with the same laws as the entire animal
series" he supports by a number of very ill-founded arguments — for in-
stance, he makes the human liver undergo a crayfish and a mollusc stage —
but also on reasoning which has been accepted by modern supporters of the
theory. This doctrine, which had already been sweepingly rejected by Rudol-
phi, has certainly been very widely debated in our own day, but at all events
it has had a highly stimulating effect upon research; its subsequent fate will
be discussed later on. The theoretical conception that Meckel thus formed
he applies in detail in the special part of his work wherein, with a many-
sided and at the same time thorough knowledge of his subject, which but
few had so far emulated, he describes the structure of the organic systems
throughout the animal kingdom. With extreme thoroughness he discusses
and compares the bone-structure in the vertebrates, describing especially
the bones of fishes in great detail and making new discoveries in this latter
MODERN BIOLOGY 359
field; the musculature and digestive canal, the respiratory and circulatory-
systems, are also carefully described. This radical detailed knowledge of
Meckel's has had a considerable influence on the development of biological
science. Moreover, his general conception of the phenomena of life has like-
wise made a deep impression. His theory of evolution deserves to be men-
tioned by the side of Lamarck's; he is at all events worthy to be named by the
many who at the zenith of Darwinism sought for "pre-Darwinists," instead
of Goethe, who never discussed the problem of species. He undoubtedly had a
great influence also on biological research in Germany during the succeeding
era; the very word "BiUungsgeserz" sounds quite familiar to anyone who has
studied Haeckel's works for instance, in which the word '"law" occurs so
often in passages where "hypothesis" or "theory" should have been written
instead. Nor can there be any doubt that his penchant for bold comparisons
and derivations has not been without its influence on the modern school,
which has made the derivation of the organs of the higher animals from
corresponding less developed forms the chief aim of biology. This morpho-
genetical school has largely applied Meckel's ideas, although employing an
entirely different standard of criticism, so that Meckel, who so essentially
belongs to romantic natural philosophy, stands as an intermediary between
this school of research and that which included a man like Gegenbaur
amongst its notable members.
Comparative anatomy in France
While thus the exact biological method in Germany only gradually suc-
ceeded in getting rid of its connexion with natural philosophy, the same
method in France had no difficulty in triumphing over the more speculative
method of natural research represented by Lamarck and Geoff'roy; it was
Cuvier and Bichat who with their pioneer work determined the direction in
which the scientists of the next generation were to follow with fair una-
nimity. Thanks, then, to these precursors, France acquired before other Euro-
pean countries a science of life-phenomena free from the romantic infusions
of speculative philosophy, but, on the other hand, this science eventually
became extremely conservative, and when the theory of the origin of species
appeared, French research repudiated it with greater vehemence than any
other. Of these French biologists from the beginning of the century some
practised a purely experimental method; these will be dealt with below. As
a direct pupil of Cuvier and Bichat, however, Blainville is worthy of mention,
for he furthered the cause of comparative anatomy long and successfully.
Henri Marie Ducrotay de Blainville was born in 1777 at Arques, in
Normandy, of noble family. He was brought up at a monastic school, and
when it was closed down during the Revolution, he went to Paris, where he
first of all applied himself to painting, apparently with but little enthusiasm
or success. A mere chance — a lecture by Cuvier that he happened to go to
360 THE HISTORY OF BIOLOGY
hear — aroused his interest in biology; at about the age of thirty he became
a pupil of Cuvier's, quickly obtained, through his recommendation, a post as
assistant, and eventually, after having held certain other appointments,
became his master's successor. By that time, however, the relations between
them had long been severed, for the pupil's quick-tempered and unreasonable
disposition could not reconcile itself to the calm considerateness of the mas-
ter. Blainville himself, however, was a splendid teacher, attracting large
audiences and devoting himself with indefatigable energy to his teaching
and research work up to a ripe old age. He died in 1850.
Blainville was a biologist with many and varied interests. Amongst his
works the Manuel d'actinologie et de xpophytologie is worthy of mention, in
which he gives the results of his thorough investigations into the lowest
animal forms, and further an Osfeograplne, which deserves to be coupled with
Cuvier's investigations into the present-day and fossilized vertebrate ani-
mals. His theoretical conception of biology he has recorded in three works:
De r organisation, des animaux, Cours de physiologie, and Histoire des sciences
de I'organisme. In these he presents a view of life-phenomena that is in many
respects original and has proved of value for the future. The first work is
introduced with a survey of the objects and methods of comparative anat-
omy. First of all an account is given of vegetable and animal chemistry,
wherein, curiously enough, the universally accepted contrasts between the
alimental process of plants and animals are considered doubtful. Then the
animal is characterized as a "combination of certain organs, which give rise
to certain forces — inter alia, a digestive and a motive force — assuming a
definite form and influencing external surroundings in a definite manner."
As methods of getting to know the structure of the animal are adduced ob-
servation, experiment, and a logical mode of thought, after which are named
certain pioneers in this sphere, among whom one seeks in vain for the name
of Cuvier — a characteristic touch, showing the pupil's bitter feelings
towards his master.
Blainville' s theory of cellular tissue
Next, an account is given of the structure of the animal body — this forming
one of the most brilliant sections of the work. Here Blainville declares with
a confidence such as was never shown previously that the cellular tissue is
the fundamental substance of the animal organism, the element which is
formed earliest and out of which all the organs are evolved. Of this tissue
it is said that it represents the finest and most extensive element in the ani-
mal body and that it is formed of thin membranes, which cross one another,
so that cystic interstices arise. As modifications of this tissue are mentioned
the skin, the mucous membranes, connective tissue, bone, vascular systems,
and finally — the most complex of all — muscle and nerves. Regarded from the
modern point of view, Blainville's conception of the cellular tissue as the basis
of the animal body is certainly imperfect, but when compared with the cellu-
MODERN BIOLOGY 361
lar-tissue theories of his predecessors, Caspar Friedrich WolfF and Bichat, it
represents an undoubted advance; it is one of many examples of how knowl-
edge in a given sphere progresses, as it were, fumblingly from generation to
generation until, finally, the decisive word has been spoken. In regard to the
conception of tissues and the part they play in the formation of the organs,
Blainville otherwise associates himself closely with Bichat, and he likewise
adopts the latter's theory of organic and animal life, which, however, he
employs with considerably greater moderation than its creator. For the rest,
he believes a living body to be a kind of chemical workshop wherein fresh
molecules are constantly being conveyed and old ones removed, where the
combination is never fixed, but always, so to speak, "in nisu," resulting in
constant motion and heat. This view of life is certainly not vitalistic, but
Blainville nevertheless emphasizes in what follows the contrast betwxen
"general" and "vital" forces, both unknown as to their real nature, but the
former far more measurable than the latter; both operate in the living body
and life's intensity is dependent upon the ascendancy of the life-forces over
the general forces. In this sphere, then, Blainville wavers somewhat between
divergent principles, and on the whole he has been counted amongst the
vitalists. The two primary qualities of life are, according to him, "cowpo-
sition" and " dkom-position" ; in the former is included the absorption of
nourishment, in the latter not only excretion, but also reproduction. Among
the alimental organs are also counted the organs of motion and, in general,
everything that moves the external bodily form, to which Blainville as-
cribes a fundamental importance for all knowledge of animal life. His sys-
tem rests entirely on this basis and thereby acquires a somewhat artificial
character; nevertheless, it has done considerable service, chiefly in the fact
that here for the first time a definite line of demarcation is drawn between
amphibians and reptiles, which all subsequent research has confirmed. For
the purposes of his special presentation of comparative anatomy Blainville
prefers to go from the higher form to the lower, his arguments in justifica-
tion of which are somewhat reminiscent of Lamarck's "degradation" theory.
In other respects, too, his treatment of comparative anatomy is based on a
form of theoretical speculation that renders the actual method of presenting
his subject highly artificial; it would, however, take too long to go more
deeply into these questions. Undoubtedly Blainville's works contain, besides
much that is absurd, a number of ideas of immense value, both in detail and
as a whole. Among these may be specially mentioned the importance he at-
taches to the stages of embryologic development as a basis of comparison
between the animal forms, a principle that, as is well known, has since
proved of fundamental importance to comparative anatomy. For this fact
science has to thank a number of works in the sphere of embryology that
were brought out during the period now under discussion. To this sphere,
therefore, we shall now proceed.
CHAPTER V
THE PROGRESS OF EMBRYOLOGY
Ovists and animalculists
THE MOST IMPORTANT FEATURES of the earlier history of embryology
have already been referred to in previous sections of this work — how
even Hippocrates had observed the development of the hen's Qgg, how
Aristotle studied the embryology of various animals, how Fabricius,
Harvey, Malpighi, and C. F. Wolff each in turn made valuable contributions
to the knowledge of the development of the embryo, especially in the hen's
egg, which had remained throughout the most easily available object of
investigation, but also in a number of other animals, chiefly, of course,
mammals. These inquiries were naturally much influenced by the speculations
on the process of development that succeeded one another during different
epochs; in this respect, the "preformation" theory, which prevailed for a
time, had a most unfavourable effect, seeing that its champions, for obvious
reasons, cared but little for practical observations of the embryonic develop-
ment — everything having been ready-formed from the beginning, there
was, of course, no need for observation. This explains why the eighteenth
century, during which the preformation theory held sway, proved so barren
in embryological observations; instead of investigating, scientists wasted
their time in profitless speculations and controversies between ovists and
animalculists. Some of the latter certainly reached the height of absurdity
when they saw in the spermatozoa the true agents of reproduction, with the
consequence that they succeeded in distinguishing under the microscope in
every human spermium, with the aid of their imaginations, a complete
miniature human being with all the limbs ready formed. It was not until
the close of that century that embryology received a fresh impetus; C. F.
Wolff made a beginning with his, certainly exaggerated, epigenesis theory
and his embryological observations based thereon; Cuvier, who was in-
terested in all biological problems, also made weighty contributions to this
subject; Blainville has just been mentioned as a promoter of embryological
research; nevertheless, science has mainly to thank certain German scientists
for its most important progress in this direction, progress which, in fact,
gave rise to an essentially new view of life-phenomena. As has often hap-
pened with pregnant problems in the history of science, this, too, was dealt
with simultaneously by several observers, each of whom contributed his
MODERN BIOLOGY 363
portion towards its solution. In the following we shall deal with Pander,
who investigated the germ layers in the embryo of the hen; Rathke, who dis-
covered the branchial slits in the embryo and the circulation in conjunction
therewith; and, in another connexion, Purkinje, who discovered the germ-
cell in the hen's egg. The first place among the creators of modern embryol-
ogy, however, is held by von Baer, one of the great personalities in the field
of research in the nineteenth century.
Karl Ernst von Baer was born in 1791 on the Piep estate in Esthonia,
the son of a landowner belonging to the German nobility of that country.
Upon leaving school at Reval he matriculated at the recently founded Uni-
versity of Dorpat, where he applied himself to medicine. He continued his
studies in branches of that subject in Vienna, but from there, realizing that
he was not made for a doctor, he proceeded to Wurzburg in order to be trained
as a theoretical scientist. The teacher of anatomy in that university at the
time was Ignaz Dollinger (1770-1841), a disciple of Schelling's, who,
combining his master's passion for philosophical speculation with a radical
knowledge of anatomy, especially interested himself and his pupils in prob-
lems of evolutional history. It was here that von Baer's research took the
course in which he was eventually to go further than any of his contempo-
raries. After completing his studies he was appointed professor at Konigs-
berg and carried out his principal investigations at that place. In 1834 he
accepted an invitation to become an academician at St. Petersburg. There his
activities won a brilliant success, and honours were lavished upon him ac-
cordingly. The scientific works of his old age, however, do not possess the
same importance as those of his early years. This is due primarily to the fact
that he to a great extent divided his interests; upon official request he under-
took several journeys to diff'erent parts of the Russian Empire and as a result
became interested in a number of different problems — anthropology, ethnog-
raphy, archasology, and even etymology. Having been allowed to resign his
post, he settled at Dorpat and died there in 1876. He was honoured by his
countrymen in many ways; the Esthonian nobility published at their own ex-
pense a splendid edition de luxe of the autobiography that he had written in
his old age, and after his death a bronze statue was raised to him at Dorpat.
Von Baer discovers the egg of mammals
There can be no doubt that von Baer won his greatest fame through the
embryological works written in his youth. He published the results of these
in a brochure entitled De ovi mammalium genesi, which came out in 18x7, and
a larger work, IJber Entivicklungsgeschicbte der Tiere, of the years 1818 and 1837.
In the first-mentioned treatise he describes the most important of the dis-
coveries he made in this field — namely, the egg of mammals in the ovary.
Apart from the vague ideas of earlier scientists on this subject, de Graaf
(Part II, p. lyz) was the first to explain at all the conditions obtaining at the
364 THE HISTORY OF BIOLOGY
earliest stages of development of mammals. He described the follicles named
after him in the ovary and believed these to be eggs; when later he discovered
eggs in the uterus of a rabbit in a later stage of growth, he supposed that
these had been moved thither from the ovary for their further development;
he met with an insoluble difficulty, however, in the fact that the further
advanced eggs in the uterus were smaller than the follicles, and, moreover,
the latter proved to be not very constant, wherefore Haller, who carefully
investigated the matter, assumed that the egg was formed out of the follicu-
lar fluid through coagulation. By carefully following the development of
the egg in dogs, von Baer learnt to know its later stages, afterwards trac-
ing its origin back by investigating a series of animals approaching nearer
and nearer to the fertilization stage. Here he found the egg to be a minute
yellowish cell inside the follicle, after which he was able to continue the
study of its progressive development.
His pioneer work on evolution
Besides these studies in mammal embryology von Baer devoted himself to
that ancient classical object of study in evolutional history — the hen's egg.
He followed its evolution with the utmost care and published the results in
the first section of the above-mentioned work tjber Entivicklungsgeschichte,
which, besides, summarizes all the then existing knowledge of the subject,
thereby becoming a pioneer work on which all subsequent research has had
to be based. The latter half of the work is a survey of the embryonic develop-
ment of all the vertebrates. Through this book von Baer has created modern
embryology, not only as an independent field of research, but also as an im-
portant branch of comparative anatomy and a means of proving the affinity
of different animal forms. In the embryo of the chicken von Baer discovered
the spinal cord, which he identified by comparison with the cord of the
selachians. He also showed in its proper light Rathke's discovery of the gill-
slits and gill-arches in the embryo. Further, he has explained the cause of the
amnion formation — a discovery comparable with the foregoing — and has
also given concise accounts of the development of the uro-genital apparatus
of the formation of the lungs, of the various stages of development of the
digestive canal and the nervous system. And finally he makes his proved
ideas a basis for a general evolutional theory, which, it is true, contains a
mass of natural-philosophical notions, but on the other hand gives a clear
survey of the connexion in evolution and far excels all previous theoretical
representations, although, since it is ignorant of the part played by the cells
in the generative process of the organism, it cannot be called modern. For
the process of fertilization is thus substituted a vague hypothesis in which
the idea of a growth over and above the individual plays a conspicuous part:
' ' Zuerst wird die Moglichkeit eines neuen Tieres durch unmittelhares Wachsthum des
mutterlichen Korpers gegeben. Es bleibt aber nur Teil. Durch die Bejruchtung wird aus
MODERN BIOLOGY 365
dem Telle ein Ganges." This definition of fertilization is pure metaphysics,
here closely akin to Aristotle's. In opposition to Wolff's one-sided cpigenesis
theory, von Baer declares that in reality no new formation takes place in the
egg, but only transformation in the direction of increasing specialization.
True, this theory is also based on purely speculative grounds — that the
idea of the producing animal-form controls the development of the embryo —
but it at any rate leads to a result that has been accepted even in modern
times. Further, against Meckel's theory that during the embryonic stage
higher animals pass through the form of lower animals, von Baer makes one
particularly striking criticism; he maintains that no lower animals exist
that really resemble the embryonic stages of the higher animals, but, on
the other hand, the embryo of a higher animal and that of a lower animal
resemble one another more closely than do the fully developed animals. The
tissues of the embryo are less differentiated than those of the animal itself
and are therefore more like the tissues in lower animals; but a fish embryo is
from the very outset, and always remains, a fish, just as every vertebrate
animal's embryo is from the beginning a vertebrate animal. Since, then,
evolution involves a differentiation, the principle holds good that "the
more dissimilar two animal forms are, the further we have to go back in
evolutional history to find an agreement." The common primal form for all
animals is the simple cell-form, the form of the egg and of the first embryonic
stage. Starting from these considerations, von Baer emphatically rejects the
Bonnet-Lamarckian theory of a uniform chain of development in the animal
kingdom and instead associates himself with Cuvier's type theory, which he
further develops. He maintains that a series of animals can in respect of the
development of one organ be progressive, while another organ in the same
animal series is regressive, and that one animal within a lower type may
attain to a very high development in comparison with another form which
comes low down within a higher type — the bee and the fish are cited, with
the intelligence as the standard of comparison. Each organ, therefore, should
be judged not only according to its definitive form, but also with reference
to its evolutional history; the different animal types often possess organs
having the same function, but an entirely different origin. He predicts that a
comparative investigation of the different organs in the animal kingdom on
this basis will prove of great importance for the future, and this prediction
has of course been fulfilled.
His natural philoso-phy
Side by side with these progressive ideas, and often curiously interwoven
with them, we find in von Baer a wealth of ideas of manifestly natural-
philosophical origin which must certainly seem highly grotesque to the
modern mind, but which nevertheless have undoubtedly had some influence
on the biological speculation that was to come. A number of these ideas he
3 66 THE HISTORY OF BIOLOGY
had adopted from contemporary natural philosophers, as for instance Oken's
theory of the head's being composed of vertebrae and of the jaws' having the
qualities of ribs. Others again he has certainly invented himself — for ex-
ample, the theory that the vertebrate animals are composed of a number of
tubes lying inside one another in the shape of the figure 8. He reaches the
extreme heights of Schellingism with a scheme he works out, according to
which the three tubes lying within one another are each divided into a
positive and a negative half; the epidermis, the muscles, and the nervous
membrane are thus given a plus sign, while the cutis, the bones, and the
nerve-fibres are denoted by a minus sign. Strangely enough, one comes across
fancies of this kind in many of the eminent biologists of that period; some
have already been mentioned, others will be discussed later on. It would be
quite irrational, however, to accept these confessions of the weakness of
the period for more than what they are; they are certainly striking from the
point of view of cultural history, but their significance, whether for the
activities of the scientists named or for their contribution to the general
development of science, should at any rate not be exaggerated.
Martin Heinrich Rathke may claim an eminent place by the side of
von Baer among the pioneers of embryology. He was born in 1793 at Danzig
of a wealthy burgher family. He studied at Gottingen under Blumenbach,
practised for a time as a doctor in his native town, was invited to Dorpat
in 18x9 as professor in physiology and thence, as von Baer's successor, to
Konigsberg, where he worked until his death, in i860. Being personally
a lovable character, keenly active on behalf of his science, constantly seeking
to increase his knowledge by research work at home and abroad, he was
universally esteemed by his colleagues and pupils.
Rathke's work as a biologist was many-sided and important. Among
his earliest works was an article published in a journal, "On the Develop-
ment of the Respiratory Organs in Birds and Mammals," which in point of
value may be compared with von Baer's above-mentioned embryological
treatises. It has already been pointed out that Rathke discovered the gill-
slits in the embryo of birds and mammals, as well as the ramification of
blood-vessels connected therewith. He further compared them with those of
the fishes and followed their later development — how the gill-slits dis-
appear and how the blood-vessels adapt themselves to the lungs, which are
developed from an expansion of the front part of the digestive canal. He has
also described and compared the development of the air-sacs of birds and the
larynx of birds and mammals. In another work he gives an account of the so-
called Wolffian bodies discovered by him, which he characterizes as "head
kidneys" (pronephros), and which for a time performed the function of excre-
tal organs, only to disappear later according as the true kidneys developed,
while their efferent ducts in certain animals serve as part of the sexual organs.
MODERN BIOLOGY 367
In regard to his theoretical conception of these conditions, Rathke accepts
without reservation Meckel's theory of the lower animal forms that the
higher animals pass through during their embryonic life. A more independent
theory is developed by him in a treatise tjber die rucks chre it ende Metamorphose
der Tiere, in which he records the results arrived at during his research work
in comparative embryology. Although full of difficult abstract ideas, this
article presents a view — original for the time at which it was written — ■
of a hitherto neglected phenomenon in animal life. The phrase "ruckschrei-
tende Metamorphose" is characteristic of the age. Rathke makes a cautious
reservation against the confusion of his ideas of metamorphosis with those
of Goethe; in the form in which it is presented here, it has in view the regres-
sive development which certain organs undergo during their embryonic and
early life and which concludes with their total disappearance or survival as
rudiments. Rathke cited examples of phenomena of this kind from the entire
animal kingdom, but he supports his argument mainly on examples taken
from the vertebrates, as for instance the gills and tail of the tadpole, the
Wolffian organs, etc. He declares that such organs are either dissolved or are
reabsorbed by the rest of the body, or else are knocked off and disappear; the
latter occurs if they are horny and lack blood-vessels, the former if they
possess blood-vessels by which their substance can be absorbed and made use
of in the body. And it always happens, he declares, that such a disappearance
of one organ is succeeded by the development of another which takes its
place, as for instance the lungs of the frog, which develop according as the
gills disappear, or the kidneys in the bird and mammal embryo, which take
the place of the Wolffian bodies. Only an entirely altered mode of living
during more advanced stages of development can bring about the total loss
of previously existing organs, as happens in the parasitic crustaceans. It
will at once be realized that here Rathke has shed light on one of the most
important problems of modern biology.
Rathke' s marine-zoological studies
But Rathke's activities were not merely confined to embryology; he is also
one of those who have opened up for biological research the vast field that
the seas have to offer. Cuvier was the first in modern times to draw the atten-
tion of science in this direction. Rathke, who was born and grew up in a
seaport and who in the course of his travels — in Scandinavia amongst other
countries — had his own interest in this field of research stimulated hereby,
contributed largely towards awakening it in his countrymen. In this respect
he gained much from the acquaintance he made in Bergen with the then
priest Michael Sars (1805-69), another of the pioneers of marine research,
who presented him with many valuable animal-forms and gave him much
information. Of Rathke's work in this sphere may be mentioned his careful
description of the lancet-fish — that extremely primitive vertebrate animal
368 THE HISTORY OF BIOLOGY
which has been so widely studied in modern times and which, not long be-
fore Rathke's period, had been thought to be a worm of a mollusc. Rathke's
monograph was the first detailed anatomical description of the animal and
was written with the same thoroughness that characterized his embryologi-
cal investigations. His abundant and valuable writings further contain several
monographs on crustaceans, both independent and parasitic, molluscs and
worms, as also a number of monographs on the vertebrates — for example,
on the lemming — which are worthy of mention as examples of Rathke's
extensive and radical research-work.
The third in order of the above-mentioned pioneers of embryology was
Heinrich Christian Pander (1794-1865). Born at Riga, the son of a wealthy
banker, he was able to give undivided attention to scientific work, which
had attracted him from an early age. He studied at Dorpat, Berlin, and Got-
tingen and afterwards, having made the acquaintance of von Baer, at Wiirz-
burg in the latter's company. There he carried out his pioneer work on the
development of the embryo of the chick, the results of which, thanks to
his great wealth, he was able to publish in a very fine edition. In 182.6 he
became an academician at St. Petersburg, but the very next year he resigned
and lived for some time as a landowner in the neighbourhood of Riga. In
1842. he entered the Russian Mining Board, after which he published only
works on geology.
Discovery of the germinal layers of the hen s embryo
Pander's above-mentioned treatises on the development of the hen's &gg,
which were published as early as 1817, were the fruits of work carried out
under the guidance of Dollinger and with the collaboration of von Baer.
They thus represent to a certain extent the basis on which the latter scien-
tist worked further, although there is no doubt that while they were being
written, the younger of the two friends came under the influence of the elder.
As Pander's greatest service should be recorded the fact that, taking as his
starting-point the preliminary work of Malpighi and C. F. Wolff, he dis-
tinguishes the different layers out of which the organs of the chicken embryo
are built up. These layers, which, following Wolff, he names '^Blatter'' — a
relic of the latter's attempts to compare plants and animals anatomically —
were afterwards further investigated by von Baer and have since then formed
the foundations of modern embryology. In his presentation of the continued
course of development of the embryo, however, Pander is not to be compared
with his above-mentioned contemporaries and he was unable to follow up
the promising ideas that he had produced in his early work. A work Ver-
gleichende Osteologie, which he published as an edition de luxe in collaboration
with the artist d' Alton during his visit to Germany, and which attracted
the attention of Goethe, has not the same interest as his embryological
works, and upon his return home he divided his genius between a number
MODERNBIOLOGY 3 69
of small tasks, the results of which have not attracted the attention of pos-
terity. There was created, however, by the scientists described above, a new
branch of comparative morphology, which proved of the greatest impor-
tance for the development of biology in general, in that it made possible
a far more universal and extensive study of the organ in living creatures than
had been conceivable before, embracing not only the present characteristics
of the organs, but also their evolutional history, thus proving not only a
morphological, but also a morphogenetical subject of research. From this
period we can also consider that the advent of comparative anatomy in the
modern sense dates, and its development during the succeeding decades, es-
pecially in Germany, was splendid. It was this line of research that really
dominated biological science in that country during the greater part of the
nineteenth century. But it was certainly not merely the embryological dis-
coveries that produced this fresh impetus. In other spheres, too, there opened
up for biological research, as a result of new methods and new facts, vistas
of an extent hitherto unknown. In particular, there were two methods, al-
ready previously known, it is true, but not sufficiently appreciated by the
immediately preceding generation, which were adopted at this period with
renewed interest and considerable improvements, and which produced re-
sults that fundamentally reformed the views on life-phenomena — namely,
the experimental method and microscopy.
CHAPTER VI
THE DEVELOPMENT OF EXPERIMENTAL RESEARCH AND ITS
APPLICATION TO COMPARATIVE BIOLOGY
The development of organic chemistry
IN THE PREVIOUS SECTION it has been pointed out that during the eighteenth
century the experimental method was applied with great success both in
animal and in vegetable biology; names such as Haller and Spallanzani,
Hales and Ingenhousz are sufficient proof of this. During the reign of roman-
tic natural philosophy, conditions were diff"erent; the representatives of that
school, who imagined that they could solve all the riddles of existence by
speculation, deeply scorned experiment, which they considered led to noth-
ing but fruitless artifice. Indeed, the physiological works which saw the
light during this epoch are for the most part purely speculative or else de-
voted to morphological problems. Gradually, however, reason came into
its own even in this sphere; the immense success which the experimental
method brought to contemporary physics and chemistry induced attempts
at applying that method also to biology. And this all the more so as during
the immediately preceding period eminent scientists had begun to apply
themselves with considerable success to the study of the chemical composi-
tion of living organisms. A glance at the development that had taken place
in that branch of chemistry may therefore not be out of place in this con-
nexion.
Carl Wilhelm Scheele, mentioned in the previous section as a pioneer
in gas chemistry, is worthy to be called the founder also of animal and vege-
table chemistry. German in origin and upbringing — he was born at Stral-
sund in ij^i. — in his youth he adopted the profession of apothecary in
Sweden, finally settling at Koping, a small town, where he died in 1786.
During a brief life spent in very poor circumstances he managed to carry
out unusually fruitful research work. As one part of his work it may be
mentioned that he subjected to a more thorough chemical revision than any-
one had done before him elements from the animal and vegetable kingdom;
among the results he obtained was the discovery of lactic acid, cyanuric acid,
hydrocyanic acid, and uric acid, and, further, glycerine, citric acid, and malic
acid, not to mention other equally important elements. Lavoisier, it will be
remembered, also studied phenomena in the animal and vegetable kingdoms.
A successor to him was Antoine pRANgois de Fourcroy (175 5-1 809).
370
H
P
hJ
O
«
O
>
w
Q
<
X
w
1-1
<5
H
O
o
o
z
0
(J
o
o
MODERN BIOLOGY 371
Brought up in poverty, he received financial aid for his medical studies
from Vicq d'Azyr and when still a young man became a professor of chem-
istry. He was a zealous supporter of the Revolution, was a member of the
famous committee of public safety, and eventually became director-general
of instruction. In a handbook that he wrote on chemistry, Philosophie chi-
mique, as well as in a number of other writings, he deals exhaustively with
animal chemistry; the so-called "chemical philosophy" excited consider-
able attention and was translated into several languages. In this work an
account is given of the chemical composition of plants and animals. The
essential difference between substances derived from the vegetable and the
animal kingdom is claimed to be the latter's azotic content. Vegetable ele-
ments are divided into sixteen separate substances, including gum, sugar,
fatty and fugitive oils, resin, etc. The elements derived from the animal king-
dom form three groups: albumen, lime, and fibrin; characteristic of both
the vegetable and animal kingdoms arc fermentative processes of various
kinds, whereof is described the fermenting of wine and vinegar, and putre-
faction. Besides this grouping of the components of living creatures,
Fourcroy has given us the results of a large number of valuable special
investigations into animal substances: milk, blood, gall, serum, and others.
His services have been fully acknowledged by Berzelius, who was the great
discoverer in this sphere as much as he was in chemistry in general.
Jons Jakob Berzelius was born in 1779 at Vaversunda in Ostergotland,
Sweden, the son of a poor priest. Being left an orphan at an early age, he
had to carry out his studies under severe privations, first of all earning his
living by private teaching and practising as an apothecary, and later, hav-
ing taken the degree of bachelor of medicine, in medical practice. In 1809 he
became a doctor of medicine and obtained an appointment as doctor at the
College of Surgery in Stockholm (out of which, thanks mainly to his ac-
tivities, the Carolinian Institute was eventually founded). Here he had a
laboratory in which he was able to apply himself to those chemical investi-
gations which in a few years were to bring him world-wide fame. An un-
rivalled capacity for work made it possible for him, apart from his research
and literary work, to devote himself to a large number of public duties —
thus, he became permanent secretary to the Academy of Science, whose an-
nual report he used to write in a masterly style — and he was also the recipi-
ent of innumerable honours both at home and abroad. Ill health weakened
his powers during the last years of his life. He died in 1848.
Berzelius' s animal chemistry
Berzelius's work as the creator of the science of chemistry in general is
universally known, and falls, moreover, outside the scope of this history.
In actual fact, he mastered the whole of chemistry as no one else has ever
done since his time, and he created something new in all the spheres in which
37i THE HISTORY OF BIOLOGY
he worked. One of his most important contributions, however, is his inves-
tigation of the substances that are produced by life on the earth. In his Lec-
tures on Animal Chemistry, published during the years 1806-8, he expounds
his conception of life-phenomena in general and, besides, records a number
of fresh facts in the sphere of organic analysis, which he later still further
augmented. These purely chemical investigations into the composition of
blood, gall, milk, bone, fat, and many other elements, really belong to the
science of chemistry rather than to biology; although they exercised a fun-
damental influence on the knowledge of life and the functions of the living
body during the succeeding period, a detailed account of their results would
hardly be in place here, but the views on the phenomena of life which Ber-
zelius formed as a result of these investigations have naturally played an
important part, both in his own time and in the age that followed; an ac-
count of them is therefore of interest, quite apart from the interest always
excited by ideas expressed by one of the great pioneering personalities of
the world.
As sources from which he derived his view of the phenomena of life
Berzelius mentions first of all the works of Bichat and Fourcroy; in partic-
ular, the former's explanation of the tissues of the body and their functions
formed the basis of his entire conception of life-phenomena. Reil, too, con-
tributed largely thereto, while Fourcroy's role was primarily that of a pre-
cursor in the purely chemical sphere. What chiefly distinguishes Berzelius's
general conception of nature from the conceptions of his predecessors is his
severe criticism of and opposition to any kind of "hypothesis-mongering."
While contemporary natural philosophy created brilliant thought-systems,
Berzelius introduces his "animal chemistry" with the words: "I have every-
where sought to avoid hypothesis, and where I have at any time ventured
to make insignificant guesses, they are all of such a nature that they will
soon be decided by experience. I prefer to say: 'This is entirely unknown to
us,' than to try by means of a number of probabilities to gloze over a gap
in our knowledge."
Berzelius rejects vitalism
In conformity with this principle he rejects the vitalistic theories of his
age: "Life does not lie in any extraneous essence deposited in an organic
or living body; its origin must be sought in the common fundamental forces
of primal elements, and this is a necessary consequence of the condition
wherein the elements of the living body are combined." And further on he
says: "All, therefore, that we explain with the words ' oim vitality^ is
entirely unexplained and it is an illusion if they are given any other meaning
than that of a still unknown mechanico-chemical process." Even the func-
tions of the soul are explained in the same way: "Unreasonable as it may
seem . . . nevertheless, our judgment, our memory, our reflections, as well
MODERN BIOLOGY 373
as other functions of the brain, are organic chemical processes, as well as,
for instance, those of the abdomen, the intestines, the lungs, the glands,
etc.; but here chemistry rises to a higher plane, where our spiritual research
can never reach her." Even La Mettrie himself has never expressed himself
more clearly, and though he adds in a note on materialism that "it is not
in accord with our hope and our practical innate feeling regarding the im-
mortality of soul," yet, in view of what has been said above, this makes but
little impression. It is obvious that a man like Israel Hwasser must have felt
antipathy for the man and the university whence such words originated.
When, later, it becomes a question of applying in detail the mechani-
cal theory in life, Berzelius, like so many of his predecessors and successors,
gives way to the temptation to simplify too much, and in spite of his honest
endeavours he is unable to free himself from speculative construction. Under
the heading "Principle Components of the Animal Body" he declares that
"the phenomena of animal life are divided into two systems, the nerves and
the other organs. Life is really placed in the former, and through them the
animal lives for the moment. The latter, on the other hand, promote those
conditions whereby the animal is to live in the immediate future." The nerve
system thus represents the essential difference between organic and inorganic
nature; the plants, therefore, must also possess nerves, although they are
still unknown. This contrast between the nervous system as the real con-
servator of life and the other organs, which are called instrumental organs,
is just as unnatural as Bichat's theory of the two lives, which Berzelius
criticizes, and the idea that life cannot exist except in a nervous system is
still more unfortunate. The functions of the nervous system are, according
to Berzelius, unknown, and he utters a serious warning against adducing
electrical and other forces to explain what one does not understand. He has
only vague ideas of the structure of the brain and the nerves; he takes no
account of Malpighi's microscopical discoveries and Swedenborg's appli-
cation of them. Moreover, he makes a number of unfelicitous assertions
regarding the vascular system; he believes that the capillaries open out be-
tween the organs and that the latter grow up as a result of a stratification
of solid matter around the opening — "a kind of crystallization." In this
connexion it may be mentioned that he believes in the spontaneous genera-
tion of lower animals. On the other hand, his description of fertilization is
clear and definite; he frankly acknowledges the inability of science at that
time to explain the process, and his ideas on the development of the tgg
are extraordinarily clear, considering they were arrived at twenty years be-
fore von Baer's epoch-making discovery.
Berzelius, therefore, cannot be said to have been a very deep thinker in
the sphere of theoretical natural science, as was Galileo, for example, but
his honest and modest acknowledgment of the limitations of natural science
374 THE HISTORY OF BIOLOGY
and his opposition to any kind of hypotheses doubtless to a great extent
cleared the atmosphere in a generation that had been befogged by the fan-
tastic ideas of natural-philosophical speculation, all the more so as he cer-
tainly possessed all the authority to which a discoverer of the highest rank
in the world of natural phenomena can lay claim. His pupils were numerous
and brilliant; the most gifted chemists of the following generation received
their training from him and undoubtedly they disseminated the master's aim
to try to establish the actual phenomena in nature without any "explana-
tions by means of qualitates fere occulta ^
Experiments on live animals
The experimental research of which an account has been given above con-
cerned the chemical composition of the various organs. But the functions
of the organs as well — their vital manifestations, each separately and in
collaboration — ■ were during this period the subject of radical experimental
investigations. In this field of research Haller was a pioneer; in his foot-
steps there followed an increasing number of scientists who sought by means
of experiment on live animals — that is, vivisections — to ascertain the
course of events in animal life, both in the isolated organs and in groups
thereof, to an ever-increasing extent. These experiments led to the discovery
that the actual phenomena of life were themselves bound by laws to an extent
hitherto undreamt of; it was found that they could be made the subject
of exact research just like the chemical and physical processes in inanimate
nature. As pioneers in this sphere French and English scientists were con-
spicuous; in Germany, where formerly natural philosophy and afterwards
comparative morphology predominated, the representatives of experimental
research were also comparative morphologists, at least those of the older
generation, wherefore it is often hard to decide to which category this or
that scientist rightfully belongs.
Charles Bell was born in 1774 in the neighbourhood of Edinburgh,
the son of a country parson. Having studied in circumstances of great poverty
and taken his degree at Edinburgh University, he came as a doctor to London,
where he rapidly gained a great reputation, becoming professor of surgery
and curator of the Hunter Museum, mentioned in a previous section of this
book. In his later years he returned to his native country as professor of anat-
omy and died there in i84z. He enjoyed a universal reputation as a clever
doctor, a distinguished university tutor, and a warmly religious personality;
he was the recipient of an extraordinary number of honours. As a scientific
author he was very productive; he published a text-book on general anatomy
that gained a wide reputation, and also a large number of papers on special
subjects. In several of these latter his Christian piety is a marked feature,
particularly in an essay on the structure and function of the hand, which
represents from beginning to end a hymn of praise to the wisdom, power.
MODERNBIOLOGY 375
and love of the Creator. What chiefly made his name famous to later genera-
tions, however, was his original investigations in the sphere of nerve-
physiology. Even Galen had been aware that some nerves had to do with
motion, while others received sense-impressions, and since then the nervous
system had been investigated by innumerable scientists with increasing ac-
curacy in the matter of detail. But there still remained the question of how
the nerves running from the spinal marrow can act as intermediaries be-
tween not only motive impulses, but also sense-impressions; to this question
no answ^er had been given or else only very unsatisfactory ones. Bell recorded
his experiment on this subject in a brief paper entitled Idea of a Neiv Anat-
omy of the Brain. He had a few copies of this printed in 1811 and presented
them to his friends. In this work he describes how he severed the posterior
root of a medullary nerve without causing any muscular contraction, whereas
the act of touching the anterior root caused convulsions in the muscles. From
this he concludes that the medullary nerves have a double function, due to
their double roots. The idea, however, is hinted at rather than followed up,
and indeed this small brochure contains several similar suggestions — re-
garding the specific mental energies, a problem which J. Miiller afterwards
examined thoroughly, as well as ideas on the localizations in the great brain
and the connexion between them, an inquiry which at the time held wide
possibilities. During the next decade, however, Bell did not follow the line
he had opened up, with the result that others got in advance of him, es-
pecially Magendie in Paris. Nevertheless, Bell's work received high praise
later on, which it deserves without a doubt.
Francois Magendie was born in 1785 at Bordeaux, where his father
was a surgeon. He went to Paris to study medicine, and after studying in
great poverty and anxiety and having successfully passed his examinations, he
became prosector at the anatomical institute, then a hospital doctor, and
finally a professor at the College de France, where he worked with immense
success as a lecturer, gathering a number of distinguished pupils around him.
He systematically originated the method of ascertaining the vital phenomena
by operations performed on live animals, thereby exciting both admiration
and disgust amongst his contemporaries; the sensitive Rudolphi mentions
his experiments with horror, and his notorious ruthlessness has even gone
down to posterity. Perhaps this has been exaggerated owing to his manners
towards his fellow men; his manners were rough and his self-esteem and
scorn for the opinions of others often reached outrageous heights. But if
he was hard on others, he did not spare himself; when for the first time
cholera raged in Paris, he voluntarily undertook the task of ministering to
the sick among the poorest of the population, defying both the risk of in-
fection and the fanaticism of the populace, who believed that the doctors
had caused the disease by poisoning the drinking-water. Magendie was
376 THE HISTORY OF BIOLOGY
active both as a doctor and as a research-worker until the time of his death,
which took place in 1855.
Magendie' s criticism of contemporary vitalism
Even in his earliest writings Magendie appeared as a keen opponent of
Bichat's vitalism, and the whole of his subsequent work turns on his insist-
ence on the possibility and the necessity of applying to the phenomena of
life the laws that hold good in physics and chemistry, and, in connexion
therewith, the experimental method that brought these sciences such suc-
cess. But he also possessed from the very outset a keen eye for the limitations
of that method; he repeatedly declares that it is not possible to explain all
life-phenomena merely as physical and chemical processes. The life-mani-
festations of the nervous system in particular are called by him "vital" and
are excepted from the mode of thought that he applies to other life-processes.
These vital phenomena are to his mind inexplicable for the very reason that
the physical-chemical principles cannot be applied to them; simply to invent
on their account a number of theories of a speculative kind he considers to
be harmful; he hopes rather that in future the exact method will be appli-
cable also to as much as possible of this field of research, for that method
alone, he says, can produce results of lasting value. He would take as the
basis of his research the method that was created by Galileo and perfected
by Newton — • " to observe and to question nature by means of experiment."
Among his predecessors in the biological sphere he names first of all Borelli,
whose previously described investigations into the mechanism of animal
movements he cites with admiration and pursued still further. He applies
the same mechanical idea especially to the respiration and the circulation
of the blood, at the same time endeavouring to take as full advantage as
possible of the progress made by chemistry during his age. In doing so he
came into constant disagreement with Bichat, whose theory of the independ-
ent life-manifestations of the various organs he desired to replace as far as
possible by purely mechanical processes such as could be confirmed by ex-
periment or observation. He considers hypotheses in general to be useless;
facts alone have any scientific value and what cannot be explained with their
aid must for the time being remain unexplained. This scepticism, which he
pursues with absolute consistency, undoubtedly proved a useful counter-
balance to the unbridled speculation of preceding periods. True, even his
criticism could sometimes lead him astray, as when he accepts Spallanzani's
assertion that the spermatozoa play no part in fertilization, and yet doubts
von Baer's discovery of the egg in mammals; but on the whole his concep-
tion of nature is both sound and keen-sighted. It rests, too, on a broad basis;
although it is the vital manifestations of the human body that he studied
most carefully, nevertheless he makes constant reference to other animal
forms and he supplements his text-book on physiology with a systematic
survey of the entire animal kingdom.
MODERN BIOLOGY 377
Magendie's greatest service to biology, however, is not on the theo-
retical side, useful though his criticism of his contemporaries' hypotheti-
cal ideas v^as; his most valuable positive contribution was without doubt
the experimental technique that he created. In working it out he took
advantage of experiences gained from methods of physics and chemistry as
well as from those of surgery and internal medicine, thereby originating an
experimental procedure that even to this day forms the basis of the method
of research in physiology. He employs it in a number of important processes
in the higher animal life, especially in connexion with the phenomena of
circulation and resorption. To both these studies he successfully applied his
mechanical system of thought; to him the circulatory apparatus, as also the
respiratory system, was a mechanism, the operation of which should be cal-
culable, and indeed was partly calculated by him. With regard to resorption,
he took advantage of Dutrochet's then recent discovery of the osmotic
processes. Magendie's investigations into the nervous processes, however,
brought him the greatest fame; independently of Bell he took up the prob-
lem of the roots of the medullary nerve, studying it both experimentally and
theoretically, with far greater attention to detail than his predecessor. He
actually claimed as his own the discovery of the physiology of the sensory
and motor nerve-roots, and if by this is meant thorough investigation into
the subject, he is no doubt justified in his claim. Bell, however, appeared
once more and maintained his old claims; this led, as usual, to a not very
edifying controversy between Magendie and his predecessor. As a matter
of fact, the contrast between them, as far as regards their general concep-
tions of nature, was as wide as it could possibly be — ■ Bell looking to the
glory of God in his scientific results, and Magendie refusing to accept any
other explanation of nature than the purely mechanical. In actual fact, both
have performed considerable services in the sphere mentioned — Bell in hav-
ing been the first to determine the bearing of the problem and to establish
the different functions of the two nerve-roots in the medulla, and Magendie
in having dealt with the problem experimentally throughout and established
the fact in all its details. Bell is said to have been deterred by the painful-
ness of the experiment from pursuing it beyond establishing the above-
mentioned fact in the case of one live subject; Magendie on the other hand,
who had no qualms on that score, carried out the investigation from as
many sides as possible.
Magendie gathered round him several distinguished pupils. Among these
may be mentioned Marie Jean Pierre Flourens (i 794-1 867), who continued
his master's work in the sphere of nerve-physiology, besides making valu-
able contributions to the knowledge of the function of the skin and several
other organic systems. Chief among French physiologists, however, must
be named Claude Bernard. He was born at Saint-Julien, near the Rhone, in
378 THE HISTORY OF BIOLOGY
1 8 13, of poor peasant parents, but through the kindness of a priest was given
an opportunity of studying. For a time he served as an apothecary's appren-
tice at Lyons, afterwards trying his hand at literature, but eventually he
devoted himself to medical studies, which he completed in Paris under great
privations. He was saved from the necessity of having to seek a living as a
country doctor by Magendie, who discovered his brilliant genius and made
him his assistant. After holding a number of other posts he became his mas-
ter's successor, but in his old age exchanged that appointment for a profes-
sorship at the Jardin des Plantes. He had to carry out his experiments for a
great number of years in chilly and damp premises, with the result that he
contracted an illness that prevented him from doing any practical work for
about ten years. Instead he spent this period of his life in literary work on
subjects in the theoretical sphere, his writings being very highly thought
of. Finally he succumbed to his illness in 1878. During the last years of his
life he enjoyed a brilliant reputation; he was the recipient of many high
distinctions, both at home and abroad, and his funeral was undertaken at
the expense of the French Government. A competent judge (Chr. Loven)
declared at his death that the greatest physiologist of the age had passed
away, and subsequent generations have not challenged that judgment. And
he was no less great as a personality; he was of a warm-hearted and modest
nature, and at the same time a brilliant writer and an eloquent speaker. His
experiments were carried out in less brutal fashion than Magendie's, but
they w^ere as deeply thought out and, if possible, even richer in results than
the latter's.
Bernard's theoretical conception
As will have been seen from the above, Bernard's research work comprises
not only a series of experimental investigations, but also, during the latter
years of his life, a collection of theoretical speculations upon the phenomena
of life. He had already formulated his theoretical conceptions in their main
features in his early youth, however, and throughout his life worked for
the creation of a completely elaborated theory of life. From the beginning
he rejects as emphatically as Magendie the vitalism of the Bichat-Cuvier
school, though he is not content, like his predecessor, with a general atti-
tude of scepticism, but endeavours to analyse the problem of what life really
is. Here he arrives at the conclusion that it is not possible to define what
life is, but only to analyse its manifestations — that is, Galileo's principle.
He groups the manifestations of life under the following headings: "Or-
ganisation, Generation, Nutrition, Evolution." Of these he finds the last to be
both the most characteristic of life and the most difficult to explain from
the purely mechanical point of view; the development, out of an egg, of an
individual, all of whose parts, both large and small, are produced in regular
sequence and in definite likeness to its parents' — it is that, he thinks, which
MODERN BIOLOGY 379
mostly distinguishes living from dead matter. But, all the same, it is not to
be supposed that living creatures and their organs are, as Bichat imagined,
independent for their functions of the laws of inorganic nature; on the con-
trary, heat and cold, electricity and chemical reagents exercise a law-bound
influence on them just as they do on dead matter. But vital phenomena defi-
nitely differ from the processes of inorganic change on account of their con-
stant alternation of regeneration and dissolution, of building up and breaking
down. This relation between the vital phenomena and the general physio-
chemical conditions that govern them Bernard calls "determinism," a term
that he would substitute for "vitalism" and "materialism," both of which
he rejects. For in contrast to Magendie he considers hypotheses and theories
useful to science; they possess, it is true, little real value, but they are never-
theless inevitable, "for in every science it is impossible to proceed from a
known fact to an unknown fact without the aid of an abstract idea or the-
ory." And yet the general view of life is the business of the research-worker
himself; "no one asks whether Harvey or Haller were spiritualists or mate-
rialists; we only know that they were great physiologists, and it is their
observations and experiments that have been handed down to posterity."
Bernard thus sought to compromise between the sheer unimaginative estab-
lishing of facts, and speculation that in its efforts to create a general theory
of existence loses sight of those fundamental realities which form the vital
conditions of all natural science.
His investigations into nutrition
Bernard's fame, however, does not by any means rest primarily on the theo-
retical view of life that he propounded. It is as a practical pioneer in the
sphere of experimental biology that he has acquired so great a name. His
investigations have especially aimed at following up the process of nutri-
tion and metabolism in the animal body, and the result he attained created
in many respects an entirely new conception of them. In particular, he
showed clearly for the first time the function of the liver in the process of
digestion; he established the percentage of sugar in the liver and studied the
conditions under which this secretion takes place. Likewise, entirely new
light was thrown by him on the part played by the liver in the body's
economy in general; the liver is characterized by him as " un veritable labora-
toire vital." Whereas after the discovery of the lymphatic system the liver
was considered to be merely an organ for the preparation of bile, Bernard
found that a number of substances from the intestine are conveyed through
the cystic vein and are transformed there. Thus, thanks to him, the knowl-
edge of the absorption of nourishment in the digestive canal was placed on
a new basis. In connexion with this subject Bernard studied the production
of sugar in the human body and in animal bodies in general. He established
the fact that a stab in the medulla oblongata of an animal causes diabetes — a
380 THE HISTORY OF BIOLOGY
fact that now bears his name and laid the foundations of our knowledge of
that disease. Further, Bernard found out the function of the pancreatic juice
in the process of digestion, investigated the function of the vasomotor nerves
and the problem of heat-production in animals, and, finally, carried out a
great deal of important work in the sphere of pathology — for instance, in
regard to the effect of poisons — all contributions of the greatest signifi-
cance to the development of biology.
Whereas in France, then, the experimental method as applied to biology
was used for the purpose of finding out purely physical and chemical phe-
nomena in living creatures, in Germany the same method had a somewhat
different application; to begin with, it had to serve the purposes of the purely
speculative philosophy that was still exercising a dominating influence at
the time and was later on practised in connexion with comparative anatomy,
being aided by the use of the microscope. This co-operation had brilliant
results; a new direction was given to biology, which placed Germany in
the first rank among the centres of research in that science. We shall now
proceed to give an account of the most important of the representatives of
this school.
Johannes Evangelista Purkinje was born in 1787 at Lobkowitz, in
Bohemia, of Czech parents. His father, who was a bailiff on an estate, died
early, but through his mother's efforts the boy became a pupil at a theo-
logical college, where he learnt German and general school-subjects, and
for three years devoted himself to theology; shortly before he was to be
ordained, however, he relinquished this career and began studying philos-
ophy and medicine at Prague. His dissertation was on the subject of sight
and was influenced by Goethe's Farbenlehre, with the result that it attracted
the interest of the poet; through the latter's influence Purkinje, who had
sought in vain to procure a situation in his own country, was invited by
the Prussian Government to become professor in physiology at Breslau in
the year 182.3. '^^^ faculty had recommended another for the appointment,
so that from the beginning Purkinje found himself in a difficult position,
which was still further complicated by the fact that he was not a good lec-
turer, probably owing to his having a poor ear for German. For many years
he worked and struggled to get a physiological institute of his own, and
eventually, having overcome the opposition of his superiors and colleagues,
especially his lifelong enemy the professor in anatomy, he was able to open
the institute in 1840 — the first of its kind in Germany, and very modestly
equipped. Hitherto Purkinje had had to carry out his experiments in his
own home, the comforts of which he had for many years sacrificed to the
ends he desired to achieve. From here emerged a number of pioneer works
in various spheres of biology, performed by himself and the, in part, very
distinguished pupils he had gathered around him. After the completion of
MODERN BIOLOGY 381
the institute Purkinje's interest in science began to wane; instead he applied
himself more and more to his country's culture and politics. Even in Breslau
he appeared as an author in the Czech language, translating the poems of
Schiller and Goethe into his native tongue. Having been called to Prague in
1850, he devoted himself heart and soul to the national cause; he now spelt
his name Jan Purkyne; he worked hard for the founding of a purely Czech
university and was a member of the Young Czech party in the Bohemian
Diet. National and foreign honours were showered upon him; and, wor-
shipped by his countrymen, though at the same time hated by the Germans
in his country, he laboured indefatigably to a great old age. He died in 1869.
Purkinj".' s discoveries
PuRKiNjE is one of the great geniuses in the field of biological discovery;
a great number of facts of the highest value to our knowledge of life have
become known through him. On the other hand, he never systematically
and thoroughly investigated any particular field of inquiry, nor did he dis-
cuss theoretical problems. Even his greatest work, his investigations into
the physiology of the senses, is really only a collection of different experi-
ments and observations without any connexion other than the organ with
which they deal. His experiments on sight-physiology, the ideas for which,
as mentioned above, he obtained from Goethe's FarbenleJore, and which, in
fact, are dedicated to the poet, represent a work of fundamental importance
in their sphere. They were carried out with extraordinarily well-trained and
keen powers of observation and a corresponding gift for experiment. Visual
sensations induced by mechanical influence, by galvanic current, by various
kinds of light-impressions, are described and analysed; especially well known
are the chapters ' ' Indirektes Seben ' ' and ' ' Wabre und scbeitibare Bewegungen in
der Gesicbtssfbdre" ; famous, too, is the venous figure named after him, which
is caused by the oblique illumination of the eye. As a nicroscopist Purkinje
has likewise made remarkable discoveries, among which should be men-
tioned the germinal vesicle in the chick, discovered two years before von
Baer found the mammalian tgg; and, further, the spiral apertures of the
sweat-glands and the structure of cartilage. Of peculiar interest are the thor-
ough investigations into the existence of the cilia in the animal kingdom;
formerly these hairs were known only in protozoa and molluscs. Purkinje
discovered them and their movement in the oviduct and respiratory duct in
vertebrates — he established the movement's independence of any extra-
neous force, although, not yet being aware of the nature of the cell, he was
unable to adduce the latter's autonomous life as a cause of the phenomenon.
There is still one more of Purkinje's discoveries that deserves mention here —
namely, the axis cylinders of the nerves, and the large ramified cells in the
cerebellum, which bear his name. As a physiological chemist he became
known for his investigations into the effect of rennet on the digestive
38i THE HISTORY OF BIOLOGY
process. There are, indeed, still several of his discoveries that might well
be referred to here if space allowed.
Contemporary with Purkinje there was working in Germany a scien-
tist who, in many respects, might be called his personal antithesis, but who
who was his equal in importance for science.
Johannes Peter Muller was born in 1801 at Coblenz, on the Rhine,
the son of a shoemaker. He was of a well-to-do family and was able to indulge
his passion for study. After a brilliant career at school he went to Bonn,
where he settled down to the study of medicine. After having taken his doc-
tor's degree he spent three terms in Berlin, where he was welcomed with
paternal kindness by Rudolphi and received impressions that proved a de-
cisive factor in his further development. Having returned to Bonn, he became,
first, lecturer and afterwards, in 1830, professor at that university. When
Rudolphi died and the question of his successor arose, Muller submitted a
letter to the Prussian Minister of Education in which he applied for the ap-
pointment, at the same time drawing up an ambitious program for his future
work. He was accordingly appointed and held the professorship until his
death, in 1858. Both as a teacher and as a scientist he worked with unique
success; the circle of pupils he gathered around him has few parallels in the
history of science, as regards both results and the fame to which many of
them attained. Muller began by devoting himself to experimental and mi-
croscopical research; it was he who introduced experimental physiology into
Germany, and his services to microscopy were of no small value. During his
later years he applied himself chiefly to comparative anatomy and evolution,
and in connexion therewith to marine research, which had first been taken
up by Rathke. For this latter purpose he visited both the Mediterranean
and the Scandinavian coasts, everywhere enriching biology with his valu-
able observations. The violent exertions demanded by this many-sided
activity had, however, told upon not only his bodily powers, but also his
mind; anxieties of a practical nature also further weakened his health. He
was University Warden during the years of the Revolution of 1848 and, being
a conservative, came into repeated conflict with the revolutionary-minded
students. Some years later he was in a serious shipwreck, which cost the
life of one of his friends. These events seemed to have proved too much for
his powers. Ever since his youth his personality had been a curious com-
bination of nervous unrest, proud egotism, and deep melancholy; the last
gained the upper hand according as his worries increased and his powers
declined. One morning he was found dead in his bed without sickness's
having intervened; a common rumour, which was never contradicted, de-
clares that in despair he laid violent hands upon himself.
Johannes Miiller's scientific career may be said to be typical of that of
contemporary German biology in general — it begins in natural philosophy
MODERN BIOLOGY 383
and ends with exact research, first physiological and then comparative
anatomical. From childhood Miiller possessed a mobile and imaginative
temperament; he had a tendency to hallucinations, which he later on studied
from a scientific point of view, and the education he received at Bonn was
well adapted to develop this over-imaginative side of him. Among his
tutors, it is worth noting, were Nees von Esenbeck, the fantastic botanist,
who has previously been described, and the Schellingian Brandis. His dis-
sertation for his doctor's degree. On the Relations of Numbers in connexion with
the Movements of Animals, is also entirely in the spirit of Oken;' here one may
read, amongst other things, that "bending and stretching are the two poles
of life, the former resembling the closed bud, the latter the opened but with-
ered flower; in both night prevails, but between them moves life." In his
old age Miiller is said to have destroyed all the copies of this fantastic pro-
duction that he could lay hands on. His visit to Rudolphi distinctly cooled
his ardour for extravaganzas of this sort, while the reading of Berzelius is
said to have had an even deeper influence upon him in this respect. Before
he entirely abandoned the natural-philosophical school, however, Miiller
published his investigations into subjective sense-perceptions, which were
undoubtedly the finest work on natural science produced by German ro-
mantic philosophy. Like Purkinje, who in this subject was his predecessor,
Miiller takes as his starting-point Goethe's colour-theory. The physical
qualities of light do not interest him at all and he accepts the theory of
light's "primal phenomena (Urphanomeny although he is not blind to its
weaknesses; what attracted him to Goethe is the latter 's observations on the
subjectivity of the sense-perceptions; taking these as his starting-point,
Miiller builds up with ample material derived from personal observations
his general theory of the specific forms of energy of the sensory organs. He
establishes the fact that every sensory organ reacts in its own special way
towards every kind of irritation; for instance, the eye through light-impres-
sions reacts just as much to blows and electric current as to daylight; on the
other hand, different organs of sense react each in its own way to the same
irritation; thus, to irritation caused by electricity the eye responds through
light-impressions, the ear through sound, the tongue through taste; and
finally each sensory organ can express its individual reaction to impressions
from within, in which are produced "imaginary sense-phenomena," or what
would nowadays be called hallucinations. Through these facts Miiller has
laid the foundations of experimental sense-physiology, which has been so
diligently studied in modern times; thanks to his extraordinary powers of
observation and clearness of thought, he succeeded in doing so in spite of
the natural-philosophical principles on which his research was based. We
^ A German resume of this work is included in the 1812. number of Oken's Ish.
384 THE HISTORY OF BIOLOGY
find here, for instance, a number of statements quite in the spirit of Goethe's
speculations. The experimental method is scornfully rejected — Magendie's
experiments in particular are the object of adverse criticism — and the
search for "divine life in nature" is highly commended. The function of
physiology is said to be to comprehend the phenomena of life, not from the
point of view of experience, but from that of the idea of life. This again is a
proof that a keen observer can create fresh values in spite of a weak theoreti-
cal standpoint. Miiller's propensity for observing the life-manifestation of
his own senses was really unique, but the danger that always attends such
self-introspection threatened him no less; his nervous system was shattered
by his "fantastic sense-observations" and he fell into a state of melancholy
bordering on insanity. Rest and careful tending restored him, it is true, but
he gave up for ever these "subjective" researches and therewith also most of
the natural philosophy upon which they were based. We may say that this
mental disease involved the downfall of natural philosophy in Germany.
J. Mulle/s experiments on sensory and ?notor nerves
MiJLLER, indeed, never abandoned his idealistic view of life, but his natural
research was now based on the principles laid down by Rudolphi — a
comparative study of the phenomena of life based on the knowledge of their
organs in different animal forms. He thus took up for renewed investigation
the Bell-Magendie experiments on the sensory and motor nerve-roots and he
succeeded in finding a more suitable subject for investigation than his pre-
decessors; they had experimented on dogs and rabbits, while Miiller had re-
course to frogs, which are of a more enduring nature and can therefore lend
themselves to more careful observation. Miiller in fact essentially widened
the knowledge of these important phenomena. His reinvestigations have
thrown a special light on reflex movements. Another field of study that par-
ticularly interested him was the embryonic development of the sexual or-
gans, in which he considerably widened the field discovered by Rathke and
von Baer; he also threw considerable light on the knowledge of the evolu-
tion of the mesonephros or middle kidney. Further, he made important
observations in regard to the glandular systems of the higher animals; in
particular, he definitely determined the glands' character of closed tubes
without connexion with the blood-vessels. He embodied the whole of his
knowledge on this subject in his Handbucb der Physiologie des Menschen, the
work which contains the clearest exposition of his general biological views
and which became the authoritative source of the contemporary conception
of life-phenomena, which held good up to the advent of Darwinism.
Miiller introduces his physiology with some general observations on
the essence of life, which show how deep was the influence that natural
philosophy still had on him, even after he had broken away from it. He is a
vitalist, and a much more positive one than Bichat, for instance, who really
MODERN BIOLOGY 385
only maintained that the life-process was inexplicable by chemical and
physical methods. Miiller, on the other hand, definitely declares that there
is a special "organic creative force" that is the essential condition of life.
He points out the resemblance between his theory and that of Stahl, but
with this difference, that Stahl considered the conscious soul to be the
condition of life, whereas Miiller holds that the consciousness is something
apart from the organic creative force; the latter belongs to all living beings,
while the consciousness, "which does not create any organic products, but
only ideas," is found only in the higher animals. "This rational creative
force manifests itself in each animal in accordance with a strict law, which
the nature of every animal requires"; it exists in the embryo before its parts
are present and it produces these parts. "Der Keim ist das Ganze, Potentia;
bet der Entwicklung des Keimes entstehen die integrierenden Teile desselben Actu."
Here, apparently, Aristoteleanism recurs word for word, and it is still more
conspicuous in Miiller's constant declaration that the organization of the
living being is governed by finality. "Die organischen Kdrper unterscheiden sich
nicht bloss von den unorganischen dutch die Art ibrer Zusammensetzung aus Elemen-
ten, sondern die bestdndige Tdtigkeit, welche in der lebenden organischen Maferie
u'irkf, schafft auch in den Geset^en eines vernunffige?z Planes mit Ztveckmdssigkeit,
indem die Teile xu^n Zwecke eines Ganzen angeordnet tverden, und dies ist gerade,
was den Organismus auszeichnet ." This gives the gist of Miiller's biological
views; among the details it may further be pointed out that he emphatically
maintained the immutability of both species and genera, as well as of other
higher systematical categories in the animal and vegetable kingdom; and
again that he holds the same epigenesis theory as C. F. Wolff, whom he
greatly admired, that he believes with Rudolphi that intestinal worms are
produced by spontaneous generation, and lastly that he considers the spon-
taneous generation of the Infusoria can be neither proved nor disproved.
His vitalisffi
It need hardly be specially pointed out that this organic creative force is a
product of natural-philosophical thought; likewise, it will at once be realized
that it in no way helps to explain the course and connexion of the vital
functions. We might apply to it Galileo's above-quoted words on the
omnipotence of God as a ground for natural phenomena: that one can derive
from it anything whatsoever because it is based on no kind of necessity. And
as a characteristic consequence of this mode of thought results the idea of
finality as a law governing organic evolution. Miiller's strong insistence upon
the complete finality of the organisms is no doubt connected with his often
expressed religious respect for nature, but he arrives at no explanation of
nature in that direction; science indeed has always striven to give to its
conclusions the character of laws of necessity, but where the domination of
necessity is established, there is no room for finality; no one has commended
386 THE HISTORY OF BIOLOGY
the finality of mathematical conclusions, however useful they may have been
to science. Miiller's ideas, however, deserve all possible attention; there is
no doubt that he largely created the standard of thought which prevailed in
biological circles up to the appearance of the origin-of-species theory, and
from which the opposition to that theory largely recruited its forces. At any
rate, this legacy of Miiller's from the age of romantic natural philosophy
certainly had its influence on successive generations; it even crept into bio-
logical theories fairly effectively, although in a roundabout way, during
the greatest days of Darwinism.
The influence of Miiller's physiological text-book has been all the greater
because its special section contains information of great value based on the
results of his own original research-work; here we find a very careful and
exhaustive account of the law of specific mental energies to which we have
previously referred, and here are explained in a manner unexcelled by his
age the functions of the nervous system; here his above-mentioned investiga-
tions on that subject are summarily described. Further he declares that the
ganglion-cells of the brain perform the latter's functions, and he explains
the connexion between them. Specially noteworthy in this respect as a
summary of his results is the chapter on "Mechanik des Nervetiprinzips,"
wherein his keen powers of observation and combination, undisturbed by
any philosophical adjuncts, are very conspicuous. His exposition of the
alimental and vascular systems, as well as of the sexual organs, is very fine,
although somewhat brief.
After the "text-book" had been completed, Miiller gave up physiology.
According to his own statement, he shared Rudolphi's dislike of experi-
menting on live animals, as practised by Magendie and his school, and his
physiological works were actually based very largely on comparative ana-
tomical observations. He clearly realized that physiology could not be carried
any further in this way and he consequently went over entirely to compara-
tive anatomy, which at that time had very large fields of inquiry still un-
exploited. Miiller made a particularly happy choice when he devoted himself
to investigations into the structure of the lowest Vertebrata. Among his
works on this subject may be mentioned his monograph on the lancet-fish,
which exhaustively supplements Rathke's previously mentioned work on
that animal. But in connexion with this group special mention should be
made of his monumental work on the skeleton-system, muscles, and nerves
of the Myxinoidei, on which he spent nearly ten years. He took up for study
this subject of the most primitive group in the order of Cyclostomi because,
as he says, the boundary forms in a class are the most interesting in that they
lose a good deal of the character of the class and thereby show us the type
of the class in its most simple form. The work contains a detailed description,
exemplary in its accuracy, of the said organic systems in the Myxine glutinosa
MODERN BIOLOGY 387
and its African related types; taking that as a starting-point he makes a
detailed comparison of the skeleton, muscular and nervous systems of all
Vertebrata. It is not only the accuracy of this work that has made it a stand-
ard for the future; but the method itself — starting from special inquiry,
comparing the results with the conditions existing in the related types of
the subject and thus throwing light on the form-connexion in a wider or
narrower group of living types — was imitated during an entire period and
is to this very day by no means exhausted. It would take too long to examine
in detail the result recorded by Miiller in this work; certain it is that his
successors have had little to add to the material he investigated, while the
comparative section is also of immense value, although naturally some of its
conclusions have since been disproved. Thus, Miiller adopts the theory of
the cranium's being formed of vertebras; though he deals with the subject
more cautiously than either Oken or Goethe, his conclusions are at any rate
too far-fetched to be acceptable in modern times.
His marine research work
The result of Miiller's occupying himself with these marine animals was
that he took up with increasing interest marine research work; through his
holiday trips to Heligoland, the coasts of Scandinavia, and the Mediterra-
nean he was irresistibly attracted to the study of the life of marine animals,
which had been so very little investigated before. In this field, as also in
that of anatomy, he became a pioneer. A long series of extremely important
discoveries in the sphere of marine biology is due to him; chief among these
should be mentioned a great number of the larval forms of worms, Echino-
dermata, and molluscs, the evolution of which was found out partly by him
and partly by others who followed his example later on; further, the dis-
covery of that curious parasitical mollusc, the Entoconcha, whose origin
in the host, a holothurian, he was nevertheless unable to discover, and fur-
thermore a number of interesting observations on the life and evolution of
fishes. He is thus not only a pioneer in marine zoology but also one of the
greatest in that field that the world has ever seen. The idea of special stations
for the study of this type of life was vigorously promoted by him, while at
the same time he originated a good deal of the methodology applied in work
on the subject. If we add that Miiller also followed in Cuvier's footsteps as
a palsozoologist with great credit, we shall have given a picture, however
incomplete, of one of the most prolific scientific achievements that the his-
tory of biology has to record.
Miiller was also very distinguished as a teacher. Few biologists, if any,
have succeeded in gathering around them so many incipient scientists of the
highest rank. One of the most eminent of these has declared that the master
never taught dogmas, but only his own method. The pupils had themselves
to form their own ideas; only the method and the results achieved were
388 THE HISTORY OF BIOLOGY
common to all. This explains to a certain extent how it was that so many-
scientists of independent and original thought could be trained in this
school, men such as Schwann and Virchow, Henle, Remak, Kolliker, Du
Bois-Reymond and Helmholtz; the same circumstance may also explain the
widely differing lines of research upon which they entered, but it naturally
confirms also the extraordinary many-sidedness of the master himself. Thus,
microscopy and cytology as well as experimental physiology in its most
strictly limited sense were here developed side by side. We shall now con-
sider, to begin with, the development of the two first-mentioned branches
of research.
CHAPTER VII
MICROSCOPY AND CYTOLOGY
Improvement of the microscope
As HAS BEEN PREVIOUSLY POINTED OUT, microscopical rcseafch had a pe-
riod of brilliant success in the seventeenth century, the age of Mal-
L pighi and Leeuwenhoek. Afterwards, however, this method made
no further advance for more than a hundred years; the eighteenth century
certainly produced some microscopists of importance, such as, for instance,
Lieberkiihn, but on the whole little was achieved during this period with
the aid of magnifying apparatus. The reason for this was that the aforesaid
scientists of the seventeenth century and their contemporaries did all that
could be done with the instruments at their disposal; microscopes were and
remained imperfect, and improvements were a long time in coming. The most
serious difficulty lay in the chromatic aberration of the lenses; a colourless
object seen under the microscope would shimmer with all the colours of the
rainbow, a fact which naturally gave rise to countless misinterpretations of
the objects investigated. To procure achromatic glass, free from this fault,
was a task that occupied many scientists at that time; Newton himself de-
clared the problem to be insoluble. Eventually a Swede, Samuel Klingen-
STIERNA (1698-1765), professor of physics at Upsala, succeeded in working
out how the achromatic glass should be made, and under his instructions an
English mechanician, Dollond, constructed the first achromatic lenses. It
was some time, however, before the invention could be utilized for microscop-
ical purposes. Among those who in the beginning of the nineteenth century
constructed microscopes with achromatic lenses may be mentioned the
Frenchman Chevalier and the Italian Amici; the latter's microscopes in
particular were very fine, and there soon arose in every country microscope-
makers who produced gradually perfected instruments. The year 1817 is
named as that in which Amici demonstrated his first achromatic lens-sys-
tem and during the thirties the biological institutions, at least the more im-
portant ones, were able to obtain specimens of these improved microscopes.
It was at the beginning of that decade also that microscopical biology first
showed any notable advance, and after that the great discoveries in this field
followed one another in rapid succession.
There were two spheres in which the pioneers of the new method were
induced to try their strength; on the one hand, the structure of the higher
389
390 THE HISTORY OF BIOLOGY
animals and plants and the problem of their fundamental constituents, and
on the other hand that world of minute, independently living creatures that
the new instruments made it possible for the eye to see — in collections of
water, in infusions on parts of plants (hence Infusoria), and indeed every-
where in nature. As a result of these investigations there arises an entirely
new conception of the composition of organisms — cytology, or the knowl-
edge of cells. An attempt to show the development of this branch of knowl-
edge in summary form offers certain special difficulties; as Richard Hertwig
strikingly remarks: "The way was paved for the reform of the cell theory
through discoveries made in very different spheres and not until late in time
concentrated in a focus." We must therefore give a brief summary of these
various discoveries, though it should be mentioned in this connexion that
many important steps were taken in this field by people who otherwise
exercised little or no influence upon scientific progress. For the sake of brevity
we must confine ourselves to discussing only the most important achieve-
ments and personalities in the history of cytological research.
Works on -plant-tissues
In the beginning of the eighteen-thirties Bichat's tissue theory was still
accepted in zoology, though more or less modified by various investigators.
In botany it was different. Since the days of Malpighi and Grew it had been
known that the wood of plants is composed of cells — minute chambers
having more or less thick walls. It was a matter of dispute whether a num-
ber of other elements in the plant, especially spiral vessels and bast, were
compact or cellular. The first who attempted to compare the composition
of animal and plant was C. F. Wolff; he believed, it will be remembered,
that the construction of each represents a mass of cell-shaped forms. Among
later scientists Blainville produced a theory, mentioned in the foregoing,
that the animal organism is composed of cells, but this theory was not very
clearly developed and therefore won but little acceptance. The knowledge
of cells, however, made steady, if slow, progress, the botanists still leading
the way. To start with, it was a question of deciding whether all the parts
of the plant consist -of cells, and this led to lively discussion. Among those
who contributed to its solution may be mentioned Charles Francois
MiRBEL (1776-1854), professor of botany at the Jardin des Plantes, and
LuDOLF Christian Treviranus (1779-1864), professor at Bonn, where he
succeeded Nees von Esenbeck. Mirbel especially examined the cell-structure
in certain mosses, making valuable contributions to the subject, besides
which he resolutely maintained the cell's quality as a basis for all structures
in the vegetable kingdom. Treviranus, on the other hand, performed a signal
service in the observations he made in regard to the regular movements of
the cellular contents in a number of vegetable forms; moreover, he observed
that the spiral vessels in plants originate in cells which become stratified
MODERN BIOLOGY 39I
Upon one another and lose their intermediate walls. The scientist who is
generally mentioned as the creator of modern plant-cytology, is, however,
Hugo Mohl. He was born in 1805 at Stuttgart, of a brilliantly gifted family
in the Government service. He became a doctor of medicine and a professor,
first of physiology at Berne, then — in 1835 — of botany at Tubingen, where
he remained until his death, in 1872.. His life was typical of the modest and
reserved man of science; being unmarried, he spent his days in the laboratory,
and his evenings, after the manner of his countrymen, at a "Stammtisch"
with a few friends. Even his research work has the same quiet character;
accurate observation of phenomena, a great capacity for placing known
facts in their proper light, extremely conscientious examination of his own
ideas, and praiseworthy consideration for those of others. He was decidedly
against philosophical speculations and he never produced any summary of
his own field of research; his writings consist of a large number of short
papers. The valuable results that he achieved, however, have been acknowl-
edged by both his own and succeeding periods. During his life he received
many honours, including that of being raised to the nobility with the name
of von Mohl, and after his death his reputation was still further enhanced.
Mohl' s work on cell-re-production
Among Mohl's works should be mentioned, to begin with, his observations
of cell-reproduction. Before his time, and even later, opinions differed on
this point. He upheld clearly and convincingly that the cells in alga; and
even higher plants arise through partition-walls being formed between
previously existing cells. These partition-walls he investigated and described
vvtth great accuracy. We must, however, leave his and his contemporaries'
detailed researches in this sphere, however influential they may have been
in the development of vegetable anatomy; it need only be mentioned here
that Mohl established the cellular structure in spiral vessels, bast, bark, and
other components of plants, a point that had formerly been much debated.
Further, he carefully investigated the process of development in spores of
various cryptogams, finding therein both a confirmation and an extension of
his theory of cellular division. In another connexion we shall make reference
to some of his further important contributions to this subject. For the rest,
he was also an expert optician; a work which was unique at the time and
which is still worth reading was his Micrographie, a text-book on microscopy
and microtechnique.
Discovery of the cell-nucleus
Among the investigators who contributed to the development of cell re-
search, which was particularly active at this period, may further be men-
tioned the English botanist Robert Brown, a scientist of many parts, whose
work will be described in another connexion. Here it need only be pointed
out that it was he who, in 1831, published the discovery that to the contents
392- THE HISTORY OF BIOLOGY
of every cell there belongs, as an essential component, an "areola" or, as he
also calls it, a nucleus; this cell-component he discovered in the epidermis
of the Orchidaceas and later he established its existence in a great number of
other plant-cells. It was, however, reserved to other investigators to dis-
cover its true significance.
Cytological research was given a new direction by Matthias Jacob
ScHLEiDEN, one of the strangest scientific personalities of his age. He was
born in Hamburg in 1804, the son of an eminent doctor. He began by study-
ing jurisprudence, became a doctor of law, and took up a practice as a barris-
ter in his native town. He had, however, but little success as a pleader, a
fact that increased his naturally melancholy disposition. Finally, in a fit of
despondency he shot himself in the forehead, but without the result he in-
tended; he recovered and then resolved to devote himself to natural science. He
became doctor of both philosophy and medicine, gained a great reputation by
his writings, and in 1850 became professor of botany at Jena. After twelve
years, however, he resigned; a professorship at Dorpat, to which he was
appointed shortly afterwards, he relinquished within the year and after that
led a life of wandering, with brief sojourns in various German towns, which
lasted till his death, in 1881. The life he led fully testifies to a soul without
balance, and this is reflected in more ways than one in his scientific work.
The work that at once brought Schleiden fame was an essay in Miiller's
archives of the year 1838 entitled "Beifrage xur Phytoge^zesis.'' The question he
propounds is: How does the cell arise? Here Schleiden takes as his starting-
point Brown's above-mentioned discovery of the cell-nucleus, and his service
to science lies in the fact that he was able to appreciate its fundamental
importance, which Brown himself failed to do. From the nucleus, or, as he
calls it, the cytoblast, Schleiden sought to reconstruct the course of develop-
ment of the cell, and he made a very happy choice when he selected for the
purpose the embryonic cell as his starting-point. He made a special study of
the embryo-sac in different phanerogams, carefully examining the nuclei in the
cells in question, and discovered in them the formation that is now termed the
nucleolus or nucleal body. This discovery, however, led him to continue
the investigation along the wrong lines; he thought he had discovered that
the nucleolus is first formed through an accumulation of granulate mucus in
the uniform content of the embryo-sac and he believes it to consist of gum;
around this element is afterwards stratified the rest of the nucleus, and not
until the latter is complete is there formed on its surface a small vesicle that
grows outwards until it encloses the entire nucleus; the walls of the vesicle
thicken, and thereby the cell becomes complete. According to Schleiden,
during the further development of the cell the nucleus is in most cases dis-
solved— a statement that of course does not accord with the facts. As will
be seen, the whole of this cell-formation theory is quite out of keeping with
MODERNBIOLOGY 393
the truth, and this is still further emphasized in the eyes of a modern reader
by the fact that Schleiden uses a number of romantic-philosophical terms:
expressions such as " pofetiz^erte Zellen," "edlere Sdjte," and other similar
terms are clearly reminiscent of Goethe. What made this paper so original
is its insistence upon the independence of the cell; the plant is presented for
the first time as a community of cells, a " Poljpsfock," as it is expressly called,
and it was from this standpoint that future investigators started who with a
finer critical sense made use of the idea that Schleiden had produced.
Schleiden' s text-book on botany
There is still one more important work from the hand of Schleiden that is
worthy of mention — his Grundxiige der ivissenschaftlkhen Botanik, which was
published in i84Z and at the time created an extraordinary sensation, criti-
cism being both favourable and unfavourable. Really in its way it is a
pioneering achievement; it implies a fundamental agreement both with the
purely systematic botanical training that had hitherto been in vogue, and
with the natural-philosophical conception of the phenomena of life. In a
lengthy "methodological introduction" Schleiden propounds his general
conception of nature, which represents the most interesting part of the work.
It shows that he held a well-thought-out philosophical view of nature, ac-
quired under the guidance of Jacob Friedrich Fries, professor of philosophy
at Jena, and one of the few thinkers who during the age of romantic specu-
lation maintained an interest in Kant's mode of thought. Following him,
Schleiden declares that the aim of natural science is "to relate all physical
theories to purely mathematical grounds of explanation." With this ideal of
exact research before him he tries to convert botany into a comparative in-
vestigation of life -forms and life-manifestations, with special reference to
the evolutional phenomena in the vegetable kingdom. As the cause of all
that happens in nature, both animate and inanimate, he assumes one and
the same "form-building force"; on the other hand, he strongly denies the
existence of any special life-force, and, as had often been done before, he
refers the growth of the crystal and the organ to the same category of phe-
nomena. In spite of this he is definitely opposed to the idea of spontaneous
generation of the higher animals and even rejects Meckel's "biogenetical
principles." In a purely philosophical connexion he maintains, with Kant,
the contrast between subject and object, and consequently also between
spiritual and material entities. Schelling's and Hegel's theories on the unity
of spirit and matter he dismisses with scorn. His "free-thinking" brought
him into dispute with the theologians; at that period the latter were monists,
following Hegel, while dualism was upheld by their opponents among the
biologists; in Haeckel's time, it will be remembered, just the contrary was
the case, which fact indicates that it was really the contrast between per-
sonalities that was the essential point
/• >
394 THE HISTORY OF BIOLOGY
In the special section of the handbook Schleiden gives an explanation
of the cytology, morphology, and physiology of plants, after much the
same plan as has been followed in similar works since then. This method of
presentation really bears the stamp of genius; the actual contents, however,
offer nothing essentially new from the point of view of that age; with con-
stant and often extremely abusive criticism of the botanists of the time —
Brown and Mohl are the only ones who are let off lightly — he presents a
summary of the facts already known. Concerning the formation of the cell
he propounds his old theory, although by that time it had lost much of the
validity it formerly possessed. In fact, another scientist had entered this
field of research with an entirely new idea that eventually directed its fur-
ther line of development.
Theodor Schwann was born in 1810 in a small town in Rhenish Prus-
sia, where his father had a book-shop. He studied under his fellow-country-
man J. Miiller, at both Bonn and Berlin; having taken his doctor's degree,
he became his master's assistant. In 1839 he was called to the chair of anat-
omy at the Roman Catholic University of Louvain, and some years later to
Liege, where he worked until shortly before his death, in iSSz. He was of a
gentle and reserved disposition; he avoided polemics and therefore accepted
none of the professorships that were offered to him at German universities —
he did not like the way the German histologists quarrelled, he said — and
throughout his life he remained a devout Catholic; thus, he was in every-
thing a contrast to Schleiden, with whom nevertheless he was on friendly
terms. His scientific activities fall entirely within the period during which
he worked with Miiller; it apparently needed his master's will-power to
spur his easy-going and peaceful nature on to any exertion. As a professor he
published only some few text-books and summaries. His teaching was always
conscientiously carried out.
Schwann's work on cell-structure
Schwann's research work during hh Berlin period was both many-sided
and important. His doctor's dissertation dealt with the respiration of the
embryo of the chick; he discovered the ferment of gastric juice, to which he
gave the name "pepsin"; he studied Infusoria and experimented with fer-
mentative phenomena, which led him to deny spontaneous generation and
to declare that fermentation and putrefaction are caused by organisms. All
these works, however, fall into the shade beside that by which he established
his fame as one of the pioneers of biology — the work published in 1839 en-
titled: Mikroskopische Untersuchungen tiber die Ubereinstimjnung in der Struktur
und dem Wachstum der Tiere und Pflan':(en. He here takes as his starting-point
Schleiden's above-mentioned cell-formation theory, which he accepts in its
entirety and expands into a general theory of the basis and origin of life-
phenomena. By way of introduction he points out the fundamental difference
MODERN BIOLOGY 395
between animal and plant, which the biology of preceding ages had realized
in the fact that animals possess a vascular system, which plants lack; the
plants' "gefassloses Wachstum' was accounted for by the cell-structure or,
as it was then called, the plant's composition of independent units. Now,
Schwann had discovered in the notochord of tadpoles cells provided with
nuclei, similar to the plant-cells, and both there and in the embryonic
cartilage he believed he saw a process of cell-reproduction such as Schlciden
had described. This induced him to look for cells in all the tissues of the
animal body, and by examining these in the embryonic stage and afterwards
following their development he succeeded in establishing the fact of cell-
structure even in tissues that in a state of full growth show little or no trace
of any such structure. It is not difficult to realize the great influence that
this discovery was to have on the tissue theory, and what follows will make
it still clearer. Of still greater significance for the future, however, was his
general cell-theory, according to which, as he says, "one common principle
of evolution is laid down for the most highly differentiated elementary parts
of the organisms, and this principle of evolution is the cell-formation."
This conception of the cell as a general unit of life and as a common basis
for the vital phenomena in both the animal and the vegetable kingdom was
immediately and universally accepted; so self-evident did its truth seem to
be that it met with hardly any opposition, and in fact became the foundation
on which since then both animal and vegetable biology have developed. It
is thanks to this theory that the present age has been able to work out its
conception of life-phenomena as a connected whole; without Schwann,
Darwinism would hardly have been victorious.
In its details, however, Schwann's cell theory is very primitive; he not
only embraces Schleiden's belief in a free cell-formation out of moisture, but
takes it further. Out of moisture is concentrated, first the nucleolus, then
the nucleus, and finally the cell; this process is explicitly compared with
crystallization, and the whole concludes with reflections as to whether the
hollow form of the cell might not be accounted for by the "Imbibitions-
fdhigkeit" of its component parts — in modern terminology, its colloidal
qualities; according to him, then, the cell-formation would be a kind of
crystallization in non-crystalline elements. For the essential part of the cell
is, in Schwann's view, its hollowness; in its essence it is a space surrounded
by walls; its content is a moisture, which runs out if the wall is damaged,
while the nucleus is a transitory formation, which disappears in later stages
of development. These views were in their essentials corrected in the im-
mediately succeeding future. For the rest, Schwann made his cell-formation
theory the basis of a general theory of life, which proved to be considerably
more materialistic than that of his master, J. Miiller; as a devout Christian
he believed in the world's serving a purpose given it by the Creator, but in
396 THE HISTORY OF BIOLOGY
contrast to Miiller he found no further or greater finality in living nature
than existed in inanimate nature; the same purely mechanical forces shape
both the cell and the crystal.
Further development of the cell theory
The cell-theory which has just been described, and which has always been
called the Schleiden-Schwann theory, after its founders, was adopted and
at once followed up by other investigators, while, as mentioned above, the
two pioneers withdrew from the field. The most important contributions
made during the next few years were those of Mohl, who published a series
of new observations regarding the role of the cell in the vegetable kingdom.
In these brief but weighty papers he analyses the different components of
the cell. To him the cell is still "a vesicle formed of a fixed membrane and
containing a moisture"; the character of the membrane is the essential thing,
and the shape, consistency, and interrelation of the cellular walls are de-
scribed before anything else. Moreover, an account is also given of the con-
tents of the cell: the "viscid moisture" that forms its fundamental constituent
is carefully described; its currents, which were discovered by the Italian
CoRTi and rediscovered by Treviranus, are depicted in detail in various plant-
forms — inter alia in those, since then, classical objects of demonstration,
the Tradescantia hairs — similarly, the evolution of the cell-content is
followed through its different stages of growth, and the secondary forma-
tions that accompany it — vacuoles, chlorophyll- and starch-granules —
are described. The fundamental substance in the cell Mohl calls protoplasm;
he thereby establishes the fact that the cell-content is an element by itself
and not merely "slime" of some indeterminate kind, as Schleiden supposed.
The name, which in spite of its clumsiness has come into permanent use, is,
as a matter of fact, based on the false assumption that all the component parts
of the cell, even (and above all) the nucleus, originate in this element, the
"primal slime." The nucleus is described by Mohl in greater detail than by
his predecessors; true, it is still stated to be, as mentioned above, a deriva-
tive of protoplasm come into being through an accumulation of a granulate
substance in young cells and disappearing in the older ones, but the grossly
mechanical precipitation-theory that Schleiden and Schwann held was ac-
cepted with reserve. And, above all, division is mentioned as being the nor-
mal method of cell-reproduction; independent cell-formation is confined to
the embryo-sac alone. In regard also to the alimental physiology of the cell,
Mohl offers some interesting observations, but they must be passed over
here.
At the same time, valuable contributions to the evolution of the cell
were made by Karl Nageli, an investigator whose far-reaching activities
will be described later on. In an essay on the pollen-formation in the phan-
erogams, particularly in the Liliaceas, he describes the cell-divisions (in 1841)
MODERN BIOLOGY 397
with such care and reliability as had never been done before; even the division
of the nucleus was observed with great accuracy. His ' ' transitory cytoblasts"
are chromosomes, although, with the inferior means at his disposal at that
time, he was unable either to follow the course of development to the end
or to interpret it aright.
While, then, plant-cytology was making rapid progress, cell research
in the animal kingdom was by no means unproductive. Among those who
collaborated in the working up of this field of research it is only possible
to name a few of the most influential: to begin with, some of Johannes
Miiller's pupils, Henle, Reichert, Remak, and Kolliker.
Jacob Henle was born at Fiirth, near Nuremberg, in 1809, the son of a
Jewish merchant who later, with his entire family, adopted Christianity.
He studied at Bonn under Miiller, afterwards becoming the latter's prosector
in anatomy at Berlin. There, however, he became the victim of political
persecution; he was a liberal and a member of the Burschenschaft, with the
consequence that, like so many other youths at that time, he was arrested
by the scarified Prussian police and after lengthy law-court proceedings was
condemned for treason. His scientific reputation, however, saved him from
further rigorous treatment; Humboldt, among others, interceded for him,
with the result that he was pardoned, but he received no further appoint-
ment from the Prussian Government. In 1840 he accepted a professorship at
Zurich, somewhat later one at Heidelberg, and finally, in 1851, one at
Gottingen, where he worked until his death, in 1885.
Under Miiller's leadership Henle worked both as an anatomist and as
a biologist in the invertebrate field; afterwards he also devoted himself to
pathology. His activities as a student of cell-life are associated with a
number of special essays, and also with his Allgemeine Anatomie, an excellent
work for its period. Among his contributions in the sphere of invertebrate
research his discovery of the hair-sac mites is universally known. Best of
all his speci/lized work, however, is his investigation of the histology of
the intestinal epithelium; it was he who discovered the cylindrical epithe-
lial cells and explained the existence of the pavement and columnar epithe-
lium in the various parts of the intestinal canal. He also carefully studied
the ciliated epithelium and its distribution and it was he who created the
term "epithelium." In connexion with the intestinal mucous membrane, he
investigated the chyle vessels with great care, particularly with reference
to their terminal ramifications, which had hitherto been misinterpreted.
Henle's General Anatomy is the first histological handbook based entirely
upon cytology and undoubtedly the most original since the days of Bichat.
It begins with a chapter on animal chemistry, viewed from a contemporary
standpoint, and goes on to describe the part played by the cell as a primary
formation. His cell-theory is on the whole that of the Schleiden-Schwann
398 THE HISTORY OF BIOLOGY
school, which has previously been mentioned: the nucleus formed through
an accumulation of a granulate substance, the cell formed round the nucleus
and consisting of membrane, nucleus, and fluid content. Cell-division is
denied as far as the animal kingdom is concerned; the cell-formation is
rather compared with the emulsive phenomena that arise when oil and al-
bumen are shaken together — an attempt at an explanation which, as is
well known, has been the subject of endless variations in modern time. On
the other hand, Schwann's comparisons between cell-formation and crys-
tallization are not accepted. Henle adopts a decidedly critical attitude in
regard to speculations on the primary vital phenomena. "Explaining a
physiological fact means tracing its necessity from physical and chemical
natural laws. It is true, even these laws offer no explanation as to the ulti-
mate grounds, but they make it possible to combine a mass of details under
one point of view." On the life -force theory adopted universally by his con-
temporaries he passes the following striking judgment: "The life-force is
formally as good an explanation as the force of gravity, but it is one force
the more and this is at variance with our striving after unity."
Henle thereupon proceeds to give an account of the tissues, and, of
these, first of all the epithelial system, which indeed was best mastered
and is very well expounded. A number of other details are also excellently
explained, especially the vascular musculature, which is here for the first
time satisfactorily dealt with. In regard to the division of tissues, Henle is,
of course, far in advance of Bichat, but even his system is, from the modern
point of view, difficult of comprehension; in particular, the category nowa-
days called connective tissue is split up into a mass of sub-headings which are
often somewhat unhappily formulated. Another weak chapter is that on
the glandular system, as indeed Henle himself admits, referring to the
paucity of the investigations that have been made in that sphere. But, on
the whole, Henle's general anatomy deserves the judgment passed on it by
a later histologist who declared that it laid the foundation of modern his-
tology and on that account will survive.
Karl Bogislaus Reichert was born in 181 1 in a provincial town in East
Prussia, where his father was mayor. He studied at Konigsberg under von
Baer and in Berlin under Miiller, was called to a professorial chair first at
Dorpat, then at Breslau, and finally in Berlin, where after Miiller's death,
when the latter's professorship was divided, he took over the professorship
of anatomy, which he retained until his death, in 1883. He began his activi-
ties as a comparative anatomist with a valuable work on the development of
the gill-arches in the Vertebrata and another equally eminent work on the
embryonic formation of the frog's head. He then devoted himself to cytology
in the spirit of Schwann and applied the latter's theories to the evolution of
frog's spawn, not, it is true, without falling into the misconception prevalent
MODERN BIOLOGY 399
at the time — he believed that the separate granules in the yolk of the egg
are independent cells — but nevertheless with great accuracy in his obser-
vations of the consecutive stages of development, resulting in his establishing
the cell character of the products of division, out of which the embryo is
formed. Even the two afterwards oft-recurring expressions "BiUungs-" and
"Nahrungsdotter" originate from him. One or two works on the evolution
of the tadpole likewise contain sound observations; one of them contains a
number of general reflections on the subject of organic formation by means
of invagination, which in a certain degree foreshadowed Haeckel's gastra^a
theory. Reichert's greatest contribution, however, lies in his study of the
evolution of the connective substance; he introduced this term to imply a
number of connecting-tissue elements of different structure and has based
it upon arguments from evolutional history. In his old age Reichert was com-
pletely isolated; he refused to accept the new protoplasm theories, and still
more the origin-of-species theory, and he made no attempt to hide his disgust
when these ideas prevailed. In particular, he attacked with great vehemence
Haeckel's theory of the germ layers being homologous throughout the ani-
mal kingdom; instead, he maintained the independent origin of the separate
organs. While he was scorned by Haeckel and his contemporaries, Reichert
has to a certain extent been justified by the results of modern research, where-
on we shall have more to say in a later chapter.
Robert Remak was born at Posen in 181 5. Like Henle, he was of Jewish
extraction, but in contrast to him held to the faith of his fathers. After
studying under Miiller he became his assistant lecturer, eventually being
given the honorary title of professor, though never holding a post as or-
dinary professor. He made a living by carrying on a medical practice, and
this gradually diverted him from a scientific career. He died in 1865. His
contribution in the field of cell research is concerned partly with neurology
and partly with embryology. Thus, he discovered and described the sym-
pathetic nerve fibres called after him, and he established the fact that in the
embryonic life the nerves are constructed in the form of fibres which grow out
from the nerve-cells. He is specially worthy of remembrance, however, for the
determined opposition he made to Schwann's theory of free cell-formation;
he studied the evolution of frogs' eggs and thereby proved that the egg is a
cell that divides itself up into new cells and that this division starts from the
nuclei; he does not accept any cell-formation by accumulation in a formless
matter. Further he drew a comparison between the embryonic development
in the egg of the frog and in that of a bird: it was he who invented the terms
"holoblastic" and "meroblastic," which are still used for these two types
of egg. He distinguishes three germinal layers, which he believes to be com-
mon to the embryonic development of all vertebrates and which giwe rise,
the outermost to the nervous system, the middlemost to the musculature,
400 THE HISTORY OF BIOLOGY
and the innermost to the intestinal tube — all observations that have been
confirmed by modern research. In his later years Remak paid special atten-
tion to the study of electrotherapy, making in this sphere a valuable con-
tribution, which, however, does not belong to the history of biology.
Rudolf Albert Kolliker was born in 1817 at Zurich, the son of a
wealthy merchant. He studied zoology in his native town under the aged
Oken and afterwards went to Berlin, where, under the guidance of Miiller
and Henle, he was initiated into their method of research. When Henle went
to Zurich, Kolliker became his prosector, but in 1847 he was invited to
become professor at Wiirzburg. There he gave lectures up to 1901, when he
resigned; he died three years later. He remained a Swiss subject all through
his life. He was one of the foremost teachers of his age; many of the most
eminent biologists of the succeeding generation belonged to his school.
Kolliker's research activities lasted as long as his educational career.
He was active far into his ninth decade and was successful to the end; his
later work therefore belongs to the following epoch. As a research-worker
he was above all a microscopist; the connecting link in his work was formed
by the microscopical method, which he employed in a great number of fields
of research, everywhere with immense success, although no discovery or idea
of supreme importance attaches to his name. He gave a splendid summary of
contemporary knowledge on this subject in his Handbuch der Geivebelehre des
Menschen, published in the year i85z, which deserves to be called the first
modern histology. Its purely external form has been repeated, with the nec-
essary modifications required by the progress of science, in innumerable
text-books on this subject. Kolliker here expounds with impartiality and
far-sightedness the contemporary cell and tissue doctrine; he does not, indeed,
entirely deny free cell-formation, but he limits its existence as much as pos-
sible. The cells themselves he considers to be constructed of elemental parts :
granular and vesicular formations, to which he ascribes a certain degree of
independence in growth and development — an idea which was later adopted
by many others. He strongly insists upon the importance of the role played
by the nucleus in the life of the cell, in its multiplication by division and its
other vital manifestations. We must pass over his classification of tissues;
he did not adopt Reichert's connective-substance category, but otherwise his
classification is clear and concise and his elucidation of the structure and
vital manifestations of the different tissues is full of original observations
and excellent in its form. Kolliker produced another splendid text-book in
his Entwicklungsgeschichte des Menschen und der hoheren Tiere (1861), a summary
in clear and comprehensive form of the embryological knowledge of the time.
Kolliker s investigations
Of Kolliker's numerous original investigations it is possible to quote here
only a few of the most important, in so far as they come within the period
MODERNBIOLOGY 401
now being dealt with. Especially noteworthy is his investigation into the
spermatozoa (1841), in which he proves that they are not parasites, but a
true sexual product. Further, his fine monograph Entwkklungsgeschkhte der
Cephalopoden (1844), which he worked out in the course of a visit to Naples
and which contains an account of egg-division and embryonic development
in those animals, to which account subsequent research has had but little
to add. Of great importance, too, was his study of the smooth musculature,
the elements of which he definitely isolated for the first time, describing
them as single-celled fibrillar; their distribution in the different organs of
man and the mammals he elucidated with unprecedented completeness.
Henle, indeed, had established the musculature of the blood-vessels, but it
was Kolliker who explained its character in detail. In the sphere of neurology
he also made valuable discoveries; thus, he proved convincingly that the
nerve-fibres are connected with processes of the ganglion-cells, thereby
making important contributions to the knowledge of their structure. If we
add that Kolliker investigated with valuable results certain unicellular
animals, as, for instance, the gregarines, we shall have given some idea of
his extraordinarily many-sided research work.
There is one scientist who is worthy of mention by the side of Kolliker
— namely, Franz Leydig (1811-1905), who was a native of Wiirttemberg
and who was professor at Bonn from 1875 to 1895. As a cytologist he was
remarkable for his investigations into the invertebrates. He, too, published
a Lehrbuch der Histologic, which, remarkably enough, pays as much attention
to the tissues of the invertebrate animals as to those of the vertebrates,
thereby laying the foundations of comparative histology, which has since
been so extensively developed.
Leydig's classification of tissues is more in accordance with the modern
method than that of Kolliker; thus he groups under the heading "connec-
tive substance" not only connective tissue and cartilage, but also bone tissue,
which Reichert still kept separate. His presentation of the life and develop-
ment of the cell is likewise more modern than Kolliker's, but Leydig's book
was published four years later, and during that period cytology made great
strides year by year. A good deal of Leydig's own pioneering research-work
is recorded in this treatise; his detailed studies of the structure of the insects,
especially their digestive, glandular, and sensory organs, should be men-
tioned first of all. Leydig's other extremely conscientious microscopical in-
vestigations into worms and molluscs, as well as vertebrates, belong to the
specialized literature on those subjects; no histological specialist can aff^ord
to neglect them, but considerations of space forbid any further reference to
them here.
Cell research entered upon a new phase through the work of Rudolf
LuDwiG Carl Virchow. He was born in Pomerania in 182.1, the son of a
401 THE HISTORY OF BIOLOGY
country shopkeeper, and after finishing school applied himself to medicine.
He was a pupil of J. Miiller and after completing his studies he became as-
sistant at the Charite Hospital in Berlin, rapidly acquiring a reputation on
account of his writings on pathology and his Archiv fur pathologische Anatomic
und Physiologie, which he founded in 1847 and edited until his death. He was
sent by the Government to be a medical officer in an industrial district in
Silesia, where a serious epidemic of typhus had broken out; in the report on
his mission he represented social distress in the district as being the true
cause of the disease in such terms as created resentment in high bureaucratic
circles. When, moreover, during the revolutionary year 1848 he joined the
opposition, he was dismissed from his post. He then moved to Wiirzburg,
where he became professor in pathological anatomy and developed such bril-
liant activities in the spheres of research and education that his school soon
rivalled that of his master, Miiller. The Prussian Government recalled him
in 1856, and from that date until shortly before his death he was one of the
most brilliant personalities at the University of Berlin. He died in 1902. as
the result of an accident. He remained throughout his life faithful to his
liberal ideas; as a member of the Prussian Diet and the German Parliament
he indefatigably supported the cause of liberalism and thereby came into
constant conflict with Bismarck and the adherents of that statesman. Vir-
chow naturally had no chance against such an antagonist, and his purely
political activities were unproductive. On the other hand, his influence on
the public health services in Germany was extraordinarily effective; it was
largely due to him that the German medical system became a model for
other countries. His energy sufficed for all claims made upon it, from the
reform of the sanitary system in Berlin to the organizing of the medical corps
during the War of 1870. Above all, however, the care of the sick in Berlin
stands as a monument to his organizing genius.
Virchow' s cellular pathology
As a research-worker Virchow was really a pathologist; it was diseases and
their causes that was the chief object of his investigations. This led him
to the problem of the cells as fundamental constituents of the organism both
in health and sickness, and in the middle of the eighteen-fifties he laid the
foundations of his "cellular pathology": a theory of the cells as the true
causes of disease. When, therefore, a decade later, bacteriology began to
make headway, he refused to accept its results. An important work that he
published on tumours was never completed, and he subsequently devoted
himself, apart from politics, mostly to anthropology and archaeology. In
these spheres also he achieved much that is of value, not least on account
of his initiative — the great Museum fiir Volkerkunde in Berlin, for in-
stance, was founded by him — but this work is by no means to be compared
in importance with the products of his youth.
MODERN BIOLOGY 403
Virchow's cellular pathological-theory has been of great importance
to the development of biology, owing to the fact that he established, as
no one else had done before him, the cell's character as an independent life-
unit. He denies any form of spontaneous generation, whether within the
organism or without in nature. Just as it is impossible for an ascaris worm
to arise out of intestinal slime or an infusorian out of decaying matter, so
it is not permitted in the physiological or pathological tissue-theory for a
cell to be constructed " aus irgend einer unxelligen Substanz,." And he continues:
Wo eine Zdle entsteht, da muss eine Zelle vorausgegangen sem, ebenso ivie das Tier
nur aus dem Tiere, die Pfian^e nur aus der Pflanze entstehen kann." It is this prin-
ciple of cell multiplication, and thereby also of the cell's role in the organ-
ism as a whole, that represents Virchow's great contribution to the history
of biology. He himself applied his principle mostly to the sphere of path-
ology, in which he created with its aid a new theory of the origin not only
of tumours and other new growths, but also of purulent bodies. Otherwise,
his conception of the cell was in no way original; he mentions as its neces-
sary components the membrane and the nucleus, and considers its other
"fluid" contents to be less essential. In his general conception of the vital
phenomena Virchow is to a certain extent undecided; on the one hand, he
declares that there is a special life-force, that life is not a mechanical result
of the molecular forces of the bodily parts, while, on the other hand, he
holds that this life -force is probably of mechanical origin. Throughout his
life Virchow was of a very pugnacious disposition and used to defend his
views with great vehemence; with Haeckel in particular — once his own
pupil — he entered into violent controversies, not only on scientific, but
also on social questions, which both disputants were strongly inclined to
confuse with one another. But these disputes belong to the next era.
The modern conception of the life and component parts of the cell was
founded by Max Schultze, a man who, in spite of a short life, made a last-
ing name in the history of biology. Max Johann Sigismund Schultze was
born at Freiburg in 1815 and studied at Greifswald (his father had been pro-
fessor of anatomy in both places), and he also attended the lectures of
J. Miiller in Berlin for a short time. He was at one time a lecturer at Halle
and from there was appointed to a professorship at Bonn. He worked with
success there, but died in 1874.
Schultze's field of activities was very extensive; he devoted himself to
microscopical subjects in a number of animal classes. Specially famous are
his writings on the single-celled animals, to which further reference will be
made later on; they form one of the foundations of his cell theory. Further-
more he carried out important investigations into the microscopical anatomy
of worms and molluscs; in the Vertebrata he studied the terminal rami-
fications of the nervous system, and made weighty contributions to the
404 THE HISTORY OF BIOLOGY
knowledge of the structure of the electrical organs. All these works, val-
uable as they are, are nevertheless put in the shade by a short essay in the
Archiv fur Anatomic und Physiologic of the year 1861, entitled " Uber Muskel-
korferchen und was man cine Zellc zu nennen habe." Schultze has hereby laid the
foundations of the modern idea of the cell. "What is the most essential
thing in a cell?" he asks at the beginning of the essay. The old theory^
which, as we have seen, Virchow still embraced, would answer: " A vesicle
surrounded by a membrane, with a nucleus and fluid contents." Schultze
refers to the embryonic cells and points out that these consist of a mass of
protoplasm with nucleus, but without any surrounding walls; the membrane
which had previously been supposed to surround these cells, and which cer-
tain investigators had brought out by chemical means, he proves to be an
artificial product. He further points out that only cells without any mem-
brane can multiply by division; those cells possessing a membrane which are
found in the animal kingdom thus lead a restricted and limited existence —
"They may be likened to an incapsulated infusorian or an imprisoned ani-
mal." Again, the substance that surrounds the nuclei in the different tissues
— muscular fibrillas, connective substance — is not, as has been declared,
a substance foreign to the cell, but a transformation of the protoplasm itself.
Accordingly, the protoplasm, in conjunction with the nucleus, is the basis
of all the life-manifestations of the cell, and the very name "protoplasm,"
which had hitherto been used only by the botanists, is introduced as the
universal term for the fundamental substance in the cell. And in connexion
therewith this substance is characterized with reference to the conditions
obtaining in plants, in the lower and higher animals; it is maintained that
the cell-mass is by no means a fluid, but an element having a definite form,
a consistency which is different in different animal forms and different kinds
of cell; it is indissoluble in water and possesses, when it is free to do so, an
independent power of motion, which is characteristic for different cases. It
is sometimes possible also for a number of nuclei to be surrounded by a com-
mon protoplasm, which again can afterwards form cell-boundaries and thus
produce isolated cells. The actual word "cell" Schultze reserves for the vital
element represented by the nucleus and the protoplasm, and this meaning
has also been retained since, illogical though it is, seeing that the word
"cell" means a space ivith walls, whereas the living cell is characterized
by the fact that it lacks walls.
Improvement of histological technics
This contribution made by Schultze laid the foundations on which cell re-
search has since been built, and this marks a new era in the science of cy-
tology. The aids to research that the students of the period just described
had at their disposal had been comparatively limited — microscopes of prim-
itive construction, with which cells and tissues were studied in their natural
MODERNBIOLOGY 405
State or at the most after dissection. Just at this transition period, however,
there appears a new and far more perfected method; it was not merely that
microscopes were rapidly improved, but mechanical and chemical means of
a kind hitherto unknown were now beginning to be discovered and to be-
come widely used. Of means of preservation there were already known spirits
and certain saline solutions, which, in conjunction with boiling, were used
for the purpose of giving the objects of investigation greater durability. Now
the method was introduced of "fixing," by various means worked out for
each particular purpose, the structures that it was intended to examine. Of
these methods, chromic acid was introduced as early as in the thirties by
Jacobson, potassium bichromate at the same period by Heinrich Muller,
and osmium acid in the sixties by Schultze, not to mention a number of
similar means that have been discovered since. By the use of different colour-
ing-matter it is possible for the structures thus fixed to be brought out clearly
even in the thinnest and most transparent sections; in 1849 carmine colour-
ing was introduced by Harting, in 1863 Waldeyer's hematoxylin was pro-
duced from Campeachy wood, and in the same year Benecke's colouring
with analine associations, which, as is well known, have since then been
produced in immense numbers. In 1870 His introduced the microtome, an
instrument that can make extremely thin sections through the tissues, the
construction of which has been varied in many ways. The new discoveries
that were made possible by this methodology belonged to the next period.
CHAPTER VIII
THE CONTINUED DEVELOPMENT OF BIOLOGY UNTIL
THE ADVENT OF DARWINISM
I . Experimental Rearch Work
Development of organic chemistry
WHILE BIOLOGICAL RESEARCH was yielding the abundant results which
have been described above, it was subject to very important in-
fluences from other natural sciences in two special spheres. We
have described how, thanks to Berzelius, chemistry had extended its inquiries
to the sphere of living beings, and how an immense number of substances of
quite a peculiar kind were analysed and described. These substances, which
nowhere exist in inanimate nature and might consequently appear exclusively
to have "life" to thank for their origin, were called organic associations;
their existence was considered to be one of the most palpable proofs that
life itself was in its essence utterly distinct from the phenomena that take
place in inanimate nature, and even independent of the chemical and phys-
ical laws that govern lifeless matter. Organic chemistry thus became, the
more it developed, the strongest support for the theory of a special life-
force as the essential precondition for all that takes place in animate nature.
The theories maintaining this force therefore gained ground amongst an ever-
increasing number of biologists; as we have seen, Johannes Miiller embraced
a theory of this nature, as also did many of his school, and even a scientist
like Magendie, opposed to speculation though he was, could not help ac-
knowledging the invalidity of the ordinary chemical laws when applied to
living nature. It was in these circumstances that Wohler made his great con-
tribution to natural science.
Friedrich Wohler was born in 1800 near Frankfurt am Main; he be-
came a doctor, but after taking his degree he devoted himself entirely to
chemistry. In order to obtain the best training available at the time he went
to Berzelius and worked in his laboratory for a year under the strict control
of the master. Having returned home, he became a teacher at a Geiverbe-
schule in Berlin and eventually professor at Gottingen, where he died in i88i.
He was a very distinguished student of chemistry, but his other activities
are overshadowed by his synthesis of urea out of cyanammonium. By this
406
MODERNBIOLOGY 4°7
discovery an element of pronounced '•vital- character had been produced
out of components that in their turn may be entirely produced out of simple,
inoreanic elements. This first synthesis of an organic element out of inor-
ganic components was naturally succeeded by countless others; organic chem-
fstry which at one time had been thought to embrace elements produced by
life and impossible to arrive at in any other way, thus became a chemistry of
the carbon^compounds, the unique character of which is due to the nature
of the elements with which it operates, but which otherwise has recourse
entirely to the methods and theories of general chemistry. The science ot
chemistry has thus become a knowledge of phenomena that are governed
throughout nature by the same laws, with the result that a new possibility
arose of combining separate phenomena under one common point of view.
Indestructibility of energy
Op still more radical importance, however, was another discovery which
was made at a somewhat later date than that just mentioned and which led
to the well-known law of the indestructibility of energy. In earlier times
Leat was regarded as an element, a kind of 'fluid," like electricity and
although Lavoisier proved its imponderability, both he and his pupils re-
tained the ancient idea as to its essence. Gradually, however, attention was
once more attracted to the fact, known ever since ancient times, that heat
arises through friction, whence conclusions were drawn as to the connexion
between heat and mechanical action. The law-bound condition that arises
therein was elucidated in the forties by several investigators working on
the subject simultaneously and independently of one another - which only
proves how ripe for solution the problem really was. Although the phenome-
non falls entirely within the sphere of physics, there were, strangely enough
two scientists with an essentially biological training who played a deci ive
part in its solution, and this fact, as well as the impottance of the subject
in itself, justifies our going into it in somewhat greater detail.
Touus Robert Mayer was born in .814 at Heilbronn the son of a well-
to-do apothecary. He studied medicine and, having passed his examinations
became a practitioner in his native town. He was seized, however, with a
d re to see something of the world and he succeeded in obtaining an ap-
ponment as a doctor on a Dutch vessel sailing to Java. Having returned
home, he once more settled down as a doctor in Heilbronn and died there
'° ' Du'ting hTs'stay in Java Mayer had noticed that in venesection the blood
in the veins was of a far lighter colour than that in Europe. He star ed ou
to discover the reason for this and finally came to the conclusion that the
metabolism in the body is dependent on the temperature "f/^e "mosphe e,
the warmer the temperature, the weaker the conversion of subs ance is re-
qu ird to be in order that the body may perform its normal functions
4o8 THE HISTORY OF BIOLOGY
retaining its normal heat. But then there must also be a definite ratio between
the heat produced by combustion in the body and the work that the body
performs during a given period, or, in still more general terms, a certain
amount of heat must correspond to a certain amount of work. Upon his re-
turn home Mayer published in Liebig's Annalen der Chemie in 1841 an essay
in which he expounded his theory, and in connexion therewith the method,
which is still in use, of calculating the dynamical equivalent of heat when
the unit of heat represents the amount it takes to heat up a given quantity
of water one degree, and the unit of work represents the force required to
lift a given weight to a given height — in our days one kilogram one metre.
For this ratio he gave a number, which, however, was later found to be in-
correct. Shortly after the publication of Mayer's report the English physi-
cist J. P. Joule published a theory based on years of experiment and having
the same gist as Mayer's, but giving a more correct number to represent the
heat equivalent; moreover, it was founded on more substantial proofs and
supported by a greater number of facts. Then in the year 1847 came out
Helmholtz's essay Von der Erhaltung der Kraft, in which the law of the in-
destructibility of energy was elucidated from all points of view and was given
its mathematical formula. During the succeeding years, however, Mayer had
further elaborated his theory; of particular value to biology was his essay
Die organische Beivegung in ihrem Zusammenhange mit dem Stojfwechsel, which was
printed in 1845 as a pamphlet because it was refused by the editors of scien-
tific journals. In this essay he applies the law of the indestructibility of en-
ergy to the vital phenomena in the animal and vegetable kingdoms, gives
an account of the mutual relation between muscular action and the digestion
in the body's exertion of energy, and at the same time shows the process of
assimilation in plants to be the foundation of life on the earth, and solar
energy to be its ultimate source. In consequence of this he feels it to be
superfluous to assume a special life-force as a source of the metabolism in
the living body. This caused but little feeling of satisfaction amongst the
biologists of his age; as a matter of fact, the whole theory of the nature of
force met with opposition even on the part of the older physicists. When
this theory eventually won the day, Mayer considered that due attention
had not been paid to his right of priority. This wounded his naturally sen-
sitive feelings, which were exposed to still more serious shocks in the year
of the revolution, 1848; he was, in fact, from both a political and a religious
point of view, strictly conservative, with the result that he joined a differ-
ent camp from that of the majority of natural scientists, who were for the
most part liberals, and besides he fell out with his brothers, who took part
in the revolution. As a result of all these vicissitudes his nerves were com-
pletely shattered; finally, after an attempt at suicide, he had to be placed
under restraint, and in accordance with the custom of the time he was put
MODERN BIOLOGY 409
into a strait waistcoat. By degrees, however, he recovered his health and
in his old age had the satisfaction of being universally recognized as the one
who had first laid down the principle of the conservation of energy.
Of Mayer's rivals Joule belongs entirely to the history of physics. Helm-
holtz, on the other hand, worked both as a physicist and as a biologist
and therefore deserves further mention in this place. Hermann Ludwig Fer-
dinand Helmholtz was born in i8ii at Potsdam, where his father was a
teacher in the gymnasium. He studied medicine in Berlin, where he was one
of J. Miiller's pupils, and became first of all an army doctor, afterwards being
appointed professor of physiology at Konigsberg (in 1849), ^^'^ later holding
the same appointment at Bonn and Heidelberg. In 1871, however, he was
made professor of physics at the University of Berlin, and somewhat later
he became director of a newly-founded physico-technical institute at Char-
lottenburg. These two posts he held until his death, in 1894. Being univer-
sally regarded as one of the foremost scientists of his day, he was the recipient
of innumerable honours both at home and abroad. His research activities
were also as multifarious as any that natural science has had to record in
recent times. As his career testifies, he was an expert in both biology and
physics; besides this, he was not only a mathematician and a philosopher
of high standing, but also an excellent stylist and an eloquent speaker. As
his doctor's dissertation he published a valuable account of the nerve-cells
in ganglia and the nervous ramifications emanating therefrom in different
animal forms. His measurement of the rapidity of the reproduction of im-
pressions through nerve-fibres was of fundamental importance. Of still
greater significance, however, was his work as a sense-physiologist. Modern
physiological optics in particular were in all essentials founded by him. He
invented the ophthalmoscope, by the aid of which it has become possible for
doctors to examine the retina of the eye; he further explained the mechanism
of lens-accommodation and also founded the theory of colours and colour-
perceptions that has been adopted in modern times. Physiological acoustics
were likewise founded by him; he explained the connexion and mechanical
action of the bones of the ear, as also the part played by the organ of Corti
in the perception of tone quality. Again, from the point of view of purely
theoretical science, he worked out a theory of the sense-perceptions, in
which he dealt with such abstract and complicated questions as the rela-
tion of the geometrical quantities to the conception of sense, and the justi-
fication of the geometrical principles based thereon. His purely physical
and mathematical work naturally falls outside the scope of this history.
Helmholtz works out the theory of indestructibility
The above-mentioned paper tjher die Erhaltung der Kraft (On the Conservation
of Forced, nevertheless, deserves further mention here. As a result of it, the
law of the conservation of energy was given the theoretical formula that
4IO THE HISTORY OF BIOLOGY
has been adopted ever since. Helmholtz, who was physicist, mathematician,
and biologist, had, in fact, special qualifications for laying down on an em-
pirical basis an exact formula for this generally accepted principle, and dur-
ing the succeeding epoch it was his name that was most often associated
with this radical change in the general conception of nature. It was this very
contribution, however, that caused Helmholtz a good deal of unpleasant-
ness. In his paper he had made no mention of Mayer, because he was not
aware of his first essay, but in a later lecture he fully acknowledged Mayer's
priority. This, however, was not enough to satisfy Mayer's admirers; one of
them, E. Diihring, a lecturer in philosophy, made an extremely bitter and per-
sonal attack on Helmholtz, as if he had sought to appropriate an honour to
which he had no right. Disciplinary action was taken against Diihring, but
Helmholtz kept silent until Mayer, broken in health, had passed away; then
he took up the challenge and pointed out how Mayer, with all due acknowl-
edgment both to his genius and to his right of priority, had nevertheless
based his views on speculation rather than on empirical research. On this
point Helmholtz was undoubtedly right; Mayer was no experimental scien-
tist — he never had a laboratory at his disposal — but he was a brilliant
thinker who, with the aid of the observations of others and his own ideas,
achieved his epoch-making results by theoretical means. The history of
natural science proves, however, that theoretical conclusions of this kind are
seldom given the same significance as conclusions drawn from the student's
own empirical observations — Swedenborg's brilliant scientific speculation,
which, though the work of a genius, was nevertheless forgotten by the im-
mediately succeeding generations, is of course the classical example of this —
and besides through his ill health Mayer was prevented from following up
his idea, which he would undoubtedly have done had circumstances per-
mitted. As it was, he had to divide the honour with Joule, the experimenter,
and Helmholtz, the universally trained thinker and observer; all three con-
tributed towards working out the principle of the conservation of energy —
certainly the most important theoretical contribution of the past century in
the field of natural science.
Through the principle of the conservation of energy the experimental
study of living organisms received a powerful stimulus; immediate steps were
taken to apply to as many life-phenomena as possible this new conception,
which placed all phenomena in existence, both animate and inanimate, in
one single simple and clear causal connexion, and which offered the hope of
being able to bring all manifestations of life, even the most complex, under
the same simple explanatory principles that physics and chemistry had al-
ready adopted. The following decades were therefore a period of brilliant
achievement in the field of experimental physiology; both its aims and its
methods took definite form during that period, so that in the medical fac-
MODERN BIOLOGY 411
ulties at the universities this science likewise has its own representatives and
its own laboratories provided with special equipment. One or two of the
most important representatives of this line of research will be cited here as
examples showing the aims of physiology during this period and its attempts
to realize them.
Emil du Bois-Reymond was born in 1818 in Berlin. His parents were of
French extraction and came from Neuchatel, which then belonged to Prus-
sia. Some time after their son's birth they moved back to their home district,
so that the boy grew up in a French environment, but at the same time,
thanks to family influence — his father was a Prussian official — he ac-
quired a strong affection for Prussia. After completing his school studies,
therefore, he went to the University of Berlin, where, after some wavering
as to a career, he applied himself to medical studies and became a pupil of
J. Miiller. In 1858 he became Miiller's successor as professor of physiology
and held this post until his death, in 1896. He never had any very large cir-
cle of pupils; but the influence on the educated public which he exercised
as secretary to the Berlin Academy of Science was all the greater. The lec-
tures that he had to hold annually in this capacity proved to be brilliantly
eloquent; he usually took some subject from the theory of history or natu-
ral science, sometimes even discussing political questions of the day, for
Du Bois-Reymond was, in spite of his French mother-tongue, a warm Ger-
man, or rather Prussian, patriot, with an almost devout reverence for the
reigning family. These lectures displayed deep scientific general knowledge
and keenness of thought and they possessed a lasting value in German lit-
erature.
Electric currents in the living body
In 1840 Du Bois-Reymond was commissioned by J. Miiller to study the phe-
nomena of electric currents in the nervous and muscular systems and he was
thus led to take up a field of research that he never afterwards abandoned.
He recorded his results in an important work entitled Untersucbungen tiber
tierische Electrizitdt, the first part of which came out in 1848, hut the work
was never completed; the last part was published in 1884. Au R. Tigerstedt
has justly remarked, it is seldom that an investigator has for so long occupied
himself exclusively with so limited a sphere of research. That Du Bois-Rey-
mond must nevertheless be counted amongst the pioneering natural scien-
tists of his age is due to the general principles he expressed and consistently
applied in his research work. In the foreword to his great book he expounds
his ideas on the innermost essence of vital phenomena, and pf-obably the
weaker sides of the vitalistic theory have never before and seldom since
been subject to such keen and striking criticism as here. The old life-force
is reviewed from all sides and the arguments in its favour are refuted one
after another; the finality of the living organisms, which so impressed
411 THE HISTORY OF BIOLOGY
J. Miiller, is rejected in view of the equal finality prevailing in the inanimate
universe; the life-force's quality of resisting the chemical disintegration of
the organism — the basis of Stahl's and Bichat's systems — is likewise cast
aside in consideration of the fact that a force which without a struggle aban-
dons its material foundation is inconceivable, for force is in reality nothing
but a quality of matter; both belong to one another and together represent
an expression for natural phenomena, such as science imagines them to be.
Force and matter are ' ' von verscbiedenen Standptinkten aus auj gcnotnmene Ah-
stractionen der Dinge wte sie sind. Sie ergdnzen einander und sie setzen einander
voraus." From this he draws the bold conclusion that the difference between
organic and inorganic nature is of no importance whatever, and he finds ad-
ditional support for this assertion in the principle of the permanence of force
as formulated by Helmholtz. He maintains also that if the organism presents
phenomena which do not exist in inorganic nature, this may be due to the
fact that the elements intrinsic in them, though they may be provided with
the same qualities and none others, nevertheless enter into new connexions
with one another and therefore display new qualities. On these grounds phys-
iology should come entirely under organic physics and chemistry. His own
contribution to this plan consists in his investigations into electrical phe-
nomena in the animal kingdom. His most important discovery in this field
is the very fact that the muscles and nerves of animals during their state of
activity produce electric currents that can be observed and measured with
the aid of the usual apparatus of electro-physics. As a result of his study of
these currents he demonstrated in practice that phenomena which in the
most marked degree belonged to the manifestations of life may be dealt w4th
with quite as much exactitude as the ordinary physical phenomena, thereby
providing the most patent proofs of his theory of the physico-chemical qual-
ities of vital phenomena. He based his explanation of these electrical phe-
nomena, however, upon a theory that is untenable, a theory according to
which the muscles and nerves are composed of a kind of electrical molecules.
Generally speaking, it was a weak point in the physiologists of that era
that they overlooked the complex structural conditions of cells and tissues
and were thus tempted to deal with the phenomena of life more schemati-
cally than accords with the reality.
Besides this limited but nevertheless important specialized research, Du
Bois-Reymond, as already mentioned, contributed a number of ideas on ques-
tions of general science; specially remarkable is his lecture " Uber die Gren-
Xen des Naturerkennens" (iSyi), wherein he seeks to establish the limits of
natural research and comes to the conclusion that, though biology might
eventually master the laws governing vital phenomena as completely as
the astronomers when they calculate the motions of the heavenly bodies,
yet science would never be able to determine what matter is or what con-
MODERN BIOLOGY 413
sciousness is; in regard to both these fundamental hypotheses science must
pronounce not only an "ignoramus," but also an " ignorabhnus." While this
statement was received with unreserved approbation in many quarters —
inter alia, by such a keen-minded thinker as Albert Lange — on the other
hand, it excited feelings of extraordinary bitterness on the part of radical
students of nature, headed by Haeckel. However, this subject, as also Du
Bois-Reymond's opinions regarding the theory of origin, belongs to the
next period and will be dealt with when the time comes to describe it.
Karl Friedrich Wilhelm Ludwig appears by the side of Helmholtz
and Du Bois-Reymond as a pioneer in the field of exact physiology. He was
born at Hessen in 1816, studied at Marburg and Erlangen, and, having taken
his doctor's degree, spent some time in Berlin, where he joined J. Miiller's
circle, without, however, being included among his direct pupils. In 1846
he became professor of physiology at Zurich, and was called thence to Vienna
and later to Leipzig, where a new institute was founded for his benefit. There
he laboured for thirty years as a professor, gathering around him pupils
from all countries. He possessed rare powers of organization, which were
best displayed in the manner in which he arranged and guided his pupils'
work. In his earlier years he developed, with their assistance, a considerable
literary production; later on he preferred, modest as he was, to let his pupils
publish the ideas he suggested to them. Universally respected, he worked
with undiminished powers until the end; he died in 1895.
Ludwig's original investigations primarily concern the functions of the
vegetative organs. Thus, he has explained the connexion between the secre-
tion of the salivary glands and the nerves that affect those organs; he investi-
gated the function of the heart in detail and analysed its various phases; he
experimented with the circulation of the blood and took valuable measure-
ments of its rapidity. Furthermore, he invented the graphic method, which
has since then played an important part in physiology. In a Lehrbuch der Phy-
siologic of the year i85x he summarized his views on the phenomena of life.
Characteristically enough, this work starts with a chapter on Physiologie der
Atome," which is really a survey of animal chemistry, while the following
chapter, "Physiologie der Aggrcgatzustdnde," deals with the phenomena of dis-
solution, diffusion, and currents. Consequently the functions of the different
organs are presented from a purely physical and chemical point of view, the
author's special subjects, the phenomena of circulation and secretion, natu-
rally receiving specially radical and expert treatment. Ludwig's teaching
was, of course, conducted on similar lines; through his pupils the conception
of vital phenomena here described spread to all civilized countries.
Of radical importance for the development of biology was the fact that
the vital phenomena were thus explained by means of the same experimental
method that had been worked out earlier in physics and chemistry; it was
414 THE HISTORY OF BIOLOGY
established once and for all that the same processes that take place in inor-
ganic nature exist also in the living organism — in other words, every vital
process has its purely physico-chemical progress. As usual, however, the
great advance thus made led to an overestimation of the possibilities thereby
opened up for science; the pioneers of experimental biology, as R. Tigerstedt
justly remarks, entirely overlooked the part played by the cell and its various
structural forms in the vital processes. They saw in the living body merely
the basis for simple physical and chemical processes and they overlooked the
extremely complex structures which represent the fundamental condition for
the operation of these forces and on account of which the phenomena of life
actually become infinitely more complex than the physiologists were pre-
pared to admit. As, later on, knowledge of the cell-structure increased, there
arose in connexion therewith a mistrust of the over-simplified idea of the
phenomena of life, which in certain quarters caused a return to that vitalis-
tic biology that exact physiology imagined it had disposed of for all time.
We shall shortly make the acquaintance of this so-called neo-vitalism.
2.. Morphology and Classification
As far as comparative anatomy and morphology are concerned, the period we
are now describing is one of transition, during which the remains of the old
natural-philosophical manner of viewing life appears side by side with ideas
produced by the great discoveries that have been recorded in the foregoing.
Besides this, we find in the research work of this era much that fore-shadows
the advent of the origin-of-species theory. A review of the most important
events in the morphological research work of this period will therefore
form a suitable transition to the great epoch-making discovery of the sixties.
Contemporary with J. Miiller in Germany there appeared in England
a comparative anatomist of more than ordinarily wide significance. Richard
Owen was born in 1804 at Lancaster, the son of a merchant. At school he
gave no special indication of genius and was therefore apprenticed to an
apothecary; then, after some years, he went to Edinburgh to study medi-
cine and there obtained his doctor's degree. Having established himself in
practice in London, he devoted his leisure hours to the study of anatomy and
as a result of his work on the subject, he became an amanuensis at the Hunter
Museum, and later, after the resignation of Charles Bell, he was appointed
its director. In i860 he was made head of the natural-science department of
the British Museum and carried out its removal to South Kensington. He
held his position for an extraordinary length of time; in his eightieth year
he resigned and after that lived for some years in retirement; after failing in
health for some time he died in 1891,
Pi
<
ffl
2
o
>
H
w
1-1
w
h-)
h-1
e4
H
w
w
2;
<
X
o
MODERN BIOLOGY 415
Owen is generally counted as England's greatest comparative anato-
mist. His activities were extraordinarily many-sided, and, thanks to his
position as head of one of the world's largest museums, he had particularly
favourable opportunities for investigating a quantity of rare animal forms,
both still existing and extinct, which he described in essays illustrated with
very fine and carefully drawn pictures. Among these special investigations
may be mentioned his description of the nautilus — the first specimen that
had ever been seen of the animal itself, the shell of which, however, had
been known since antiquity, and, further, the anatomy of the Brachiopoda
and the lung-fishes. Furthermore, he made a thorough study of the gorilla
and certain other rare forms of the ape family and the curious finger-animals
from Madagascar; and amongst fossil forms, the Saurian bird, Archnsop-
teryx, and the extinct giant birds of New Zealand. His monumental work
on dental forms in the Vertebrata should also be recorded.
Homology and analogy
On the basis of this unique wealth of material he built up a number of theo-
retical speculations upon the organization of the entire animal kingdom,
which had a great influence on the biology of the succeeding age. In one of
the courses of lectures which he as curator of the Hunter Museum had to
give each year on the subject of comparative anatomy, he takes as his start-
ing-point his predecessor's plan of comparing the same organ through all
the animal groups and combines it with Cuvier's principle of examining the
mutual relationship of the different organs in one and the same animal form
in order to be able thus to ascertain the causes of the changes that the organs
have undergone in the different animal types. As a matter of fact, he adheres
to Cuvier's type theory throughout, condemning Bonnet's simple evolutional
series covering the entire animal kingdom. In making this comparison, he
proves that the same function can be exercised in different animal forms
partly by similar and partly by entirely dissimilar organs; the dragon-lizard
flies with its outstretched ribs, the flying fish and the bird with their extremi-
ties, the insects, again, with folds in the skin, which were originally gills.
This last idea, according to his own statement, he got from Oken; the gills
of the fishes and the lungs of the higher animals possess the same function,
but are not the same organs; rather the swimming-bladder and the lungs
correspond to one another, as is proved by the lung-fishes. This contrasting
relation between the function and the character of the organs he expresses
by the terms "analogy" and "homology"; "analogue" denotes "a part
or organ in one animal which has the same function as another part or organ
in a different animal," and "homologue" denotes "the same organ in dif-
ferent animals under every variety of form and function." The homologies
are of course the object of his special interest, chiefly those in the Verte-
brata, the bone-structure of which he made the subject of a special work.
4l6 THE HISTORY OF BIOLOGY
He there mentions three distinct types of homologies: special homology, or
agreement between a part or an organ and a part or an organ in another
animal; general homology, or the relation between an organ or a series of
organs and the general type in conformity with which the animal in ques-
tion is constructed; and, finally, the series homology, or what would nowa-
days be called metamerk homology — that is, the repetition of certain organs
in the same individual, segment for segment. He dilates upon these differ-
ent forms of homology, citing numerous examples. "Special" homology in
particular is discussed in detail, a uniform nomenclature, on Linnasan lines,
being given for homologous bones throughout the whole vertebrate series;
many bones that had hitherto been denoted by a prolix character here ac-
quire for the first time names of their own, besides which a number of other
names are rejected as unsuitable.
Anyone with any knowledge of modern comparative anatomy must at
once realize how important these terms and ideas created by Owen have been
to present-day biology. Indeed, the very idea of homology has proved one
of the most fertile grounds for comparative anatomy, although its real mean-
ing has become somewhat altered in the course of time. And the special
applications referred to — the comparison between swimming-bladders and
lungs, the derivation of insects' wings from respiratory organs — we find
amongst the most frequently quoted arguments on behalf of the modern the-
ory of evolution.
Owen s romanticism
Yet Owen himself was by no means a modern biologist in his general con-
ception of nature; rather, he stood considerably nearer romantic natural
philosophy. He was a great admirer of Oken, whom he extols as being a par-
ticularly deep thinker and whose theory of the cranium's being composed of
vertebra; he adopted and endeavoured to apply further; also he highly ad-
mired GeofFroy Saint-Hilaire, with whom he really has more spiritual affin-
ity than he has with Vicq d'Azyr and Cuvier. Like GeofFroy, he speculates
upon a common "archetype" for all vertebrates; he reconstructs one and
illustrates it in one of his works, and to this archetype are referred the "gen-
eral homologies" mentioned above. And he preferred not to recognize the
origin of the higher life-forms from the lower and the parallel derivation
of the more highly developed organs from the more primitive. He was ir-
reconcilably hostile to Darwin's theory in particular; upon its appearance
he challenged it anonymously and in quite a heated controversial spirit,
which certainly resulted in the exposure of a few weak points, but showed
a great lack of understanding of the true value of the theory. Later, however,
while maintaining the purely hypothetical character of the origin-of-species
theory, he acknowledged the correctness of Lamarck's assertion that only
individuals exist and that the term "species" is relative. On account of this
MODERN BIOLOGY 417
attitude, however, Owen became more and more isolated towards the end
of his life, and after his death there were erected in the museum he had
founded the statues of Darwin and Huxley, but not his own. Nevertheless,
there are to be found deep traces of his influence even in the champions of
the origin-of-species theory, chiefly perhaps in Haeckel, whose principal
work even in its very title — Generelle Morphologic — recalls one of Owen's
expressions referred to in the foregoing, and whose method of morphological
comparison shows obvious traces of the influence of the English anti-Dar-
winist.
In the same year as Owen there was born another of the foremost biol-
ogists of that period — namely, Karl Theodor Ernst von Siebold. At
the time of his birth (1804) his father was a professor at Wiirzburg, but was
afterwards called to Berlin. The son studied there under Rudolphi and at
Gottingen under Blumenbach, but after attaining his doctor's degree had
to take up a practice, first in the provinces and afterwards in Danzig, whither
he removed in order to have an opportunity of studying marine animals. On
account of his writings he was appointed professor, first of anatomy and
physiology at Erlangen, later, in succession to Purkinje, at Breslau, and
afterwards of zoology and comparative anatomy at Munich. There he died
in 1885, having for some years previously been incapable of fulfilling his
duties owing to ill health.
Besides Siebold we should mention his friend Friedrich Hermann Stan-
Nius (1808-83). Born in Hamburg, he studied at Breslau and Berlin under
J. Miiller; he became a lecturer under him and later professor at Rostock.
Thanks to his activities there, that small and ill-conducted provincial uni-
versity acquired a wide reputation. Unfortunately his career was prematurely
cut short; after some years of failing health he became the victim of an in-
curable mental disease and spent the last two decades of his life in an asylum.
The results of the collaboration of these two scientists were recorded in
a Lehrbuch der vergleichenden Anatomie, in which Siebold took the invertebrate
section and Stannius the vertebrates. This work was published in 1846 —
that is to say, about the same time as Owen's work just referred to, and
manifestly quite independently of it. A comparison between the two works
is therefore not without its interest. In both the manner of presentation is
largely the same; the organs are dealt with no longer throughout the entire
animal kingdom, but each larger main group is considered by itself, and the
organs are compared within that group; the natural-philosophical method
of comparison of the early years of the century is abandoned. The two Ger-
man biologists generally avoid all speculation; their presentation is based
solely on fact; thus, they do not possess the wealth of ideas of Owen, but
their presentation is founded on a particularly many-sided knowledge of
detail; they also pay special attention to microscopical anatomy, which
4l8 THE HISTORY OF BIOLOGY
Owen does only in a minor degree. Through their comparative anatomy
Siebold and Stannius made an extremely valuable contribution to the de-
velopment of morphological science. Each also laboured separately and pro-
duced important results.
Stannius w^as especially many-sided, both as a teacher and as an in-
vestigator; he taught pathology, physiology, and anatomy and has made
valuable additions to our knowledge in all these spheres. As a physiologist
he became famous for his attempts at underbinding the va^^ious parts of
the heart of the frog; further, he investigated the nerves of the tongue with
reference to their functions of taste and movement, and also the dependence
of muscular contraction upon nervous irritation. As a comparative anato-
mist he was especially remarkable for the splendid work he wrote on the
peripheral nervous system of the fishes, which is still authoritative, in spite
of all the more recent work accomplished in that field.
Siebold, as is evidenced by the share he took in the Comparative Anatomy,
devoted himself mostly to invertebrate research. In this sphere he was
without doubt one of the foremost of his age. With a view to furthering
the development of research in his own special sphere he founded, in asso-
ciation with Kolliker, the Zeitscbrift fiir wissenschaftUche Zoologie — the title
being chosen of set purpose in opposition to the soulless method of classi-
fication — which was started in i^^i. and has since been one of the chief
organs for the furtherance of biological science. In many specialized fields of
invertebrate research Siebold 's contributions have been considerable. Fore-
most in this regard come his investigations into parasites. It may be recalled
that even Rudolphi believed that intestinal worms arose as the result of
a diseased process in the host. At an early stage Siebold was quite convinced
that this kind of spontaneous generation could not be accepted as rational.
He adduced in proof of this argument the existence of large quantities of
eggs in the intestinal worms, which obviously indicated that these animals
reproduced themselves in the same way as other animals. But the question
that still required an answer was how the offspring of the parasites come
to harbour in a fresh host. As yet this question was insoluble. To answer
it required a knowledge of a phenomenon of evolution — the alternation
of generations — which was still unknown at the time and was elucidated
by another scientist, who must therefore be described in this connexion.
Johannes Japetus Smith Steenstrup (1813-97) was the son of a priest
from Jutland, in Denmark; he studied in Copenhagen, becoming professor
of zoology there after having been a schoolmaster for some years. He was
an extraordinarily gifted and many-sided investigator, working in many
widely differing fields of research; particularly celebrated are his investiga-
tions of peat-mosses, which he explored not only for zoological, but also
for botanical, geological, and archsological purposes. He also discovered
MODERN BIOLOGY 419
the ancient shell-mounds from the Stone Age that are called kjokkenwoddinger
(refuse-heaps), and studied them with valuable zoological and ethnographi-
cal results. He was, moreover, interested in marine research and he made a
discovery in this field that more than anything else ensured to him a place
in the history of biology — namely, the alternation of generations. A. von
Chamisso, famous as a poet and circumnavigator, had already proved that
in the Salpa free individuals and individuals bound together in chains al-
ternate with one another by generations, but this discovery had been almost
entirely neglected. Other observations of a similar kind had also been made
before, as, for instance, by Michael Sars in Norway and Sven Loven in
Sweden, and also by Siebold himself in Danzig, but it was reserved for Steen-
strup to complete the material for observation and to place it under one
common point of view. In 1841 he published his work on the alternation
of generations, in which he gives an account of the evolution of the medusas,
the Campanularia;, the Salpa, and the Trematoda, and finds the phenomenon
common to them all — namely, that there exists in them an alternation be-
tween the adolescent stages, which he terms "nurses," because without sex-
ual reproduction they develop a new generation, the sexually mature, and
this sexual stage, which, in its turn, gives rise by means of ordinary sexual
reproduction to new "nurses." Of such asexual generations he sometimes
finds many, one after another, particularly in the Trematoda, in which they
are often described as one independent genus, the cercarias. In its details this
work certainly required both correction and completion, but its service lies
in the fact that it laid down a common principle that proved to be indis-
pensable if a conception was to be formed of the evolution of a great many
of the lower animals. The fact that Steenstrup regarded these phenomena
from a strictly romantic-philosophical point of view was only in accordance
with the practice of the time; he saw in the alternation of generations a
striving on the part of nature after freedom and perfection, and on these
grounds he accounted for even the insect communities by the alternation
of generations: the sexless workers he considered to be "nurses," which
devote to the offspring a care of a more ideal character than that which the
Salpa series and the cercarias give to their own progeny.
Evolution of intestinal ivorms ascertained
It was Siebold who now succeeded in making practical use of the alternation-
of-generations idea, for he realized its fundamental importance for ascer-
taining the evolution of intestinal worms. He at once began to try to find
out by experimental means the connexion between a number of parasite
formations that had hitherto been regarded as independent of one another.
Thus he proved that the parasite which by insinuating itself in the brain
of the sheep causes the disease called "gid" or "sturdy," and which had
hitherto been described as Coenurus cerebralis, is actually an adolescent stage
4XO THE HISTORY OF BIOLOGY
of a tapeworm that lives in the intestinal canal of the dog — if dogs were
fed on coenurus-affected sheep's brains they would infallibly be infected with
this tapeworm, which is now called Tania canurus. Through the excrement
of the dog the eggs of the tapeworm are scattered about the pastures and
thus infect the sheep. In the same way another tapeworm in the dog was
traced to bladder-worms in the liver of hares, and a tapeworm in the cat
to similar formations in the rat. Again, the large and often fatal liver-para-
site echinococcus, which occurs in man, was found to give rise to an almost
microscopically small, and therefore hitherto unknown, tapeworm occurring
in the dog, the Tania echinococcus , the eggs of which are conveyed, through
too intimate contact with the dog, to the human mouth and thence to the
intestinal canal. As a result of these researches the knowledge of the intesti-
nal parasites, or helminthology, as it had already been termed previously,
was placed on entirely rational footing, and it only remained for later ob-
servers, by working along the lines laid down by Siebold, to collect fresh
facts in order to fill the gaps in the knowledge of the subject.
Likewise, as regards insect communities, Siebold produced the idea
which has since been pursued up to the present time. At first he held with
Steenstrup's attempts to explain the reproduction of the bees as a form of
alternation of generations, but he realized his mistake, chiefly through col-
laboration with a pioneer in the field of practical apiculture, the Roman
Catholic priest Johann Dzierzon (1811-1906), of Silesia. He is of course
famous as the founder of modern rational bee-keeping, and it was he, too,
who, notwithstanding his lack of anatomical training, realized before any-
one else the true relationship between the sexes in the community life of the
bees. He discovered that the queen-bee is fertilized only once in her life and
this while in flight through the air, not inside the beehive, and that the
drones are evolved out of unfertilized eggs, while workers' and queen-bees'
larvas are developed from fertilized eggs, both having female characters, al-
though in the workers they become stunted, owing to lack of nourishment.
These important discoveries, upon which Dzierzon based his reform of api-
culture, were received with strong opposition on the part of most bee-keepers
and would certainly have failed to win general acceptance had not Siebold
given them the support of his authority. Already a long while previously
(in 1837) he had carried out a careful investigation into the bees' sexual
apparatus, with the result that he had discovered the queen's receptaculum
seminis; now — at the beginning of the fifties — he placed himself definitely
on Dzierzon 's side and by means of a series of experiments and treatises on
the subject obtained victory for his views. In connexion therewith Siebold
elucidated for the first time the conditions obtaining in parthenogenesis in
insects in general, thereby introducing into biological science a field of re-
search that, especially in recent times, has aroused keen interest.
MODERN BIOLOGY 4x1
By the side of Siebold, Leuckart deserves mention as one of those who
contributed towards the progress of biology in the middle of last century.
Karl Georg Friedrich Rudolf Leuckart was born in iSiz at Helmstadt,
where his father was a business man, and studied at Gottingen, especially
under the physiologist Rudolf Wagner; he became a lecturer there, then
professor of zoology at Giessen in 1855, and was called thence to the same
chair in Leipzig in 1869. There he worked until his death, in 1898. Being
still keenly interested in the advancement of science even in his old age, he
gathered around him large numbers of pupils up to the end; helpful and
warm-hearted, original and good-humoured, he won their affection and was
universally praised after his death.
Leuckart's scientific activities were many-sided and of deep significance.
While still a young lecturer he published an epoch-making work, Uber die
Morpbologk der wirbellosen Tiere. "Descriptive zoology," he says, "must per-
mit of the same comparative, morphological treatment as anatomy." From
this point of view he discusses the zoological system prevalent at the time,
taking as his starting-point Cuvier, whose type theory he unreservedly de-
fends against the earlier belief in one single evolutional series, at the same
time upholding the idea of idealistic morphology of a fundamental form
after which "nature has constructed" the separate life-forms. Leuckart is
definitely opposed to such systematical categories as are based upon nega-
tive characters, as, for instance, the Lamarckian group of Invertebrata. Nor
indeed do Cuvier' s four types meet the requirements of comparative mor-
phology: apart from the Protozoa, whose then still undiscovered structure
did not permit of definitive morphological treatment, and the Vertebrata,
whose place was already established, Leuckart sets up five fundamental types
— namely, Coelenterata, Echinodermata, Vermes, Arthropoda, and Mol-
lusca. These have, of course, been generally accepted since then, although
with sundry modifications: Vermes have been further divided and Tunicata
have been separated from the Mollusca. Through this reform, however,
Leuckart brought the system a good step nearer the point that it has reached
today. In particular his treatment of the Coelenterata was epoch-making;
besides establishing the difference in anatomical structure between them and
the similarly radially symmetrical Echinodermata, he explains the curious
division of labour that takes place between the individuals in certain colony-
building forms within this class, principally in the group Siphonophora,
wherein the individuals in a colony are converted for the purpose of per-
forming a number of special functions necessary for the welfare of the whole
community. As a result of this elucidation of the structure of these so-called
"polymorphous" animal stocks, to a certain extent fresh light was thrown
on the term "individual" in the animal kingdom. The fact that Leuckart
believed he could compare the plants in general with these colony-formations
42-2. THE HISTORY OF BIOLOGY
does not detract very much from the value of the service he rendered in
throwing light on an important field in the biology of the lower animals.
Leuckart's discovery of micropj/le
Leuckart made another valuable contribution to the development of bi-
ology in his discovery that the thick-shelled egg of insects is invariably
provided with a canal through which fertilization takes place. This canal,
which Leuckart, associating it with a corresponding formation in the vege-
table kingdom, called "micropyle," was studied by him with great thor-
oughness in a large number of eggs of various insects. His investigation led
him to the discovery that the spermatozoa actually do penetrate into the
yolk of the egg through the canal — a discovery that essentially deepened
our knowledge of fertilization. That the spermatozoa thus play a vital part
in fertilization was a point which many investigators have had difficulty in
realizing; after all, it was not so long ago that Spallanzani's theory of the
decisive importance of the spermatic fluid had eminent supporters. J. Miiller,
it is true, had already discovered a similar canal in the eggs of the sea-urchin,
but it was Leuckart who proved its widespread existence and thereby also
its significance. Henceforth there was no doubt that the spermatozoon pro-
duced fertilization by penetrating the egg; it was then reserved for the fu-
ture to ascertain the cytological course of development.
Of Leuckart's other works should be mentioned his investigations of
the Spongida, which he referred to the Coelenterata as a result of detailed
study of their structure. Further, his experiments on the intestinal worms,
conducted in competition with Siebold; it was Leuckart who found out the
evolutional process in the two well-known human parasites Tania solium
and saginata, as also in the liver-fluke, which is so fatal to domestic animals.
It was also thanks to him that the Trichinas first became thoroughly known.
His important text-book on the human parasites is the work on which all
subsequent research in this field has been based.
A peculiar form of parasitism is presented by the crustaceans that make
fishes their hosts; in these the parasitic degeneration assumes forms such as
scarcely exist anywhere in the rest of the animal kingdom. The first real
knowledge of these animal forms was established by Alexander von Nord-
MANN. He was born in 1805 of a Germanized Finnish family in the county
of Wiborg, Finland, studied at Abo and from there went to Berlin, where
he became a pupil of Rudolphi. There he wrote his work Mikrographische
Beitrage, in which he deals with certain parasitic Trematoda, but chiefly
with the parasitic Crustacea, which were hereby brought to the knowledge
of science. The work attracted universal notice; J. Miiller in a letter speaks
of its " herrliche Beobachtungen," and on account of it the author was called
to a professorial chair in Odessa. There he applied himself to exploring the
animal world of South Russia, both recent and extinct, which he described
MODERNBIOLOGY 4^3
in a number of important works. In 1849 he was appointed professor in Hel-
singfors, where he laboured until his death, in 1866. In his old age he be-
came an original character, and nothing of his later production achieved
the fame of his early work.
The epoch now under discussion was on the whole prolific in bio-
logical students of high distinction; in the foregoing it has been possible to
name only a few of those who took a prominent part in furthering the de-
velopment of the science in particular branches; we shall now mention a
further group of important scientists who made a special study of marine
research and, above all, investigated the hitherto practically untouched
lower animal world of the ocean. Scandinavia was at the time one of the
main quarters of Europe in which interest was awakened in marine biology.
As one of its original promoters the Norwegian Michael Sars is worthy of
mention. Born in 1805, he studied theology and became a priest on the west
coast of Norway. There he began to interest himself in marine animal life
and published his observations in a couple of treatises, which attracted
widespread attention. He continued to follow the course he had thus entered
upon, and in 1854 he became professor of zoology in Christiania, where he
worked until his death, in 1869. Among his valuable contributions may
be mentioned the discovery of metamorphosis in the marine Mollusca, his
observations of the Crinoidea that are found at great ocean-depths and his
establishing their likeness to large groups of similar animals from earlier
epochs. He was a pioneer in introducing marine research into Scandinavia.
In Sweden the study of marine animal life was taken up by Sven Ludvig
LovEN. Born in Stockholm in 1809, he studied in Lund, where he took his
degree, and in Berlin under Rudolphi and Ehrenberg, made a journey of
exploration to Spitsbergen in a fishing-sloop — the first of the many voyages
of polar exploration that have started from Sweden — and finally became
curator of the zoological department of the State museum in Stockholm,
where he laboured for over fifty years. He died in 1895. He very greatly
enriched the museum's collections and founded Sweden's zoological marine
laboratory at Kristineberg.
Of Loven's observations, most frequently published in the proceedings of
the Swedish Academy of Science, may be mentioned his investigation of the
metamorphosis of the Annelida — the name "Loven's larva" still recalls
the fact — his study of the evolution of the genus Campanularia, which
provided Steenstrup with one of his ideas for the alternation-of-generations
theory, and, above all, his investigations into the evolution of the marine
Mollusca, in the course of which he established the formation of the so-
called polar bodies and their expulsion from the egg — a phenomenon the
evolutional universality of which was not determined until a long time
afterwards. He was especially interested in the embryology of the lovser
414 THE HISTORY OF BIOLOGY
animals and he carried out many valuable investigations in that field. He re-
tained his youthful interest in polar research and keenly promoted Sw^edish
voyages of polar exploration in his later years. When the theory of the glacial
period was first advanced, he embraced it with enthusiasm and produced
valuable zoological proofs of it, establishing the existence in some of the
deeper inland seas of peculiar animal forms that otherwise belong to the
fauna of the polar seas and which have manifestly survived in the lakes
since some earlier period when the sea covered the land. These surviving
forms Loven named "relicts," and since then Scandinavian research has been
occupied in their study.
The foremost Swedish biologist during this period, however, was un-
doubtedly Anders Adolf Retzius (1796-1860). Born at Lund, where his
father was a distinguished natural scientist, he studied under him and the
anatomist Florman, and later in Copenhagen under Ludvig Jacobson. When
still a young man he was appointed professor at the Veterinary Institute
in Stockholm and at the same time held a post at the Carolinian Institute,
where he at once became the greatest force the Institute had with the excep-
tion of Berzelius. This twofold work as a teacher, however, did not prevent
him from following up his biological researches both at home and on expedi-
tions, in the course of which he came into close contact with many eminent
scientists, including J. Miiller, who became a loyal friend, and Purkinje, from
whom he learned microscopical technique. He is the pioneer of comparative
anatomy in Sweden; he introduced it into the country not merely as a sub-
ject for research, but also — after overcoming strong opposition on the part
of older authorities — as a subject of medical training. His own works are ex-
traordinarily many-sided. A biographer has said of them that they are seldom
consistently worked out and that the whole of his research work was marked
by a certain restlessness. As a matter of fact, most of his literary production
consists of short articles for journals, written simply in the form of notes
without any theoretical reasoning or even observations on the earlier his-
tory of the problem under discussion. Nevertheless, many of these articles
have had a deep influence on the development of biology, owing to the great
value of the facts set forth in them, as, for instance, his account of the anat-
omy of the Myxinoidei, written in 18x2., in which these animals' vascular
and nervous systems, head-cartilage, and various other organs are described
— a work which formed the basis of J. Miiller's important monograph on
the Myxinoidei — and, further, his study of the connexion between the
spinal and the sympathetic nervous system in the horse — one of the first
of its kind, and a beautifully illustrated work. Like Purkinje, Retzius in-
vestigated the microscopical structure of dental bone, extending this inves-
tigation to a number of animal forms. In 1841 he went with J. Miiller to
the west coast of Sweden and there applied himself, inter alia, to the study
MODERNBIOLOGY 4x5
of the Amphioxus. In his later years, an ophthalmic disease having put an
end to his microscopical work, he entered upon a new field of research, in
which he undoubtedly carried out the most important work of his life —
namely, comparative anthropology. Till then this science had followed the
lines of Blumenbach; mankind had been divided into races for the most part
according to the colour of the skin. Retzius began to take an interest in
the discovery of human remains in the prehistoric graves of Scandinavia,
and in the course of his investigations found considerable variation in the
shape of the brain-cap. He extended these researches to include other human
races and thereupon found that the skulls may be divided, according to the
proportion between length and breadth, into long and short skulls — doli-
chocephalic and brachycephalic — and, further, after the shape of the facial
bones, into orthognathic and prognathic. Several peoples externally akin
were found to differ in these respects — • thus, for instance, the Germanic
are long-skulled, the Slav short-skulled — and as a result of this idea there
was created a field of human research based on true comparative anatomy,
which was afterwards followed up with splendid results by other investi-
gators: Virchow, Broca, the younger Retzius, and many others.
In France during this period marine biology made rapid progress and
contributed much of importance to the general development of our knowl-
edge in this sphere; of the students who distinguished themselves here, two
are especially worthy of mention. Henri Milne-Edwards (1800-85) ^^^
a native of Belgium, but of English descent. He came early to Paris and even-
tually worked as a professor there. Having been a pupil of Cuvier's, he car-
ried on with distinction the latter's work in the sphere of invertebrate
research. When quite a young man he published an extremely useful work,
a comparative study of the vascular and nervous systems of the Crustacea;
after that followed an extensive work on the fauna of the French coast,
in which the Annelida in particular were described with minute care. A
masterpiece of its kind is his Histoin naturelle des crustacees, wherein this
order is treated with thoroughness and perspicacity, and a system, based on
comparative anatomy, is worked out which is largely applied even today.
Further we may mention his work on the coral animals and his study of
the evolution of the ascidians, as well as a large number of deep-sea investi-
gations.
Another who worked along the same lines was Felix Joseph Henri
Lacaze-Duthiers (182.1-1901). Born in the south of France of a distinguished
family, he devoted himself to the study of medicine and became a professor,
first of all at Lille and then in Paris. His numerous works deal chiefly with
the MoUusca, the anatomy and evolution of which he worked out in detail;
amongst other things he worked out the anatomy of the purpura and its
colour-secretion, which was known to the ancients, but had since been
4^6 THE HISTORY OF BIOLOGY
forgotten. He is best known, however, as the founder of France's zoological
marine stations at RoscofF and Banyuls. On these institutions he spent large
sums out of his own purse and laid down the lines and methods on which
they were to work. Being a man of essentially conservative views, it was
only after a long time and after much hesitation that he accepted the theory
of the origin of species. He viewed with scepticism many of the movements
of his time; he had written up over the door of his laboratory the words:
"Science has neither religion nor politics" — a sentiment that certainly de-
serves greater attention than has been given to it in modern times.
3 . Microbiology
That there exists a world of organisms which owing to its small size eludes
observation with the naked eye has been known since the invention of the
microscope, but our knowledge of these creatures may be said nevertheless
to begin from the age with which we are now dealing, when an improved
microscopical technique first made possible a more thorough exploration of
this extensive field. Leeuwenhoek, the foremost microscopist of the seven-
teenth century, discovered, as previously mentioned, a number of minute
animals, partly in water taken from rivers and lakes, partly in putrefying
matter of various kinds. He studied them as carefully as he could, was con-
vinced of their character of living creatures, and declared that they multiplied
only by reproduction. During the succeeding century these investigations
went on with fresh observations in isolated cases, but without yielding any
really novel results. It was found that these minute animals exist especially
in water which has been allowed to stand over parts of plants or other simi-
lar growths, and from the fact of their existence in such "infusions" they
were called Infusoria, or infusion-animals. BufFon, in accordance with his
general theory of life, believed them to be products of the life-units existing
everywhere, while Spallanzani firmly rejected the idea of their spontaneous
generation. The scientist who first made a special study of the Infusoria,
however, was the Dane O. F. Miiller, who is therefore worthy of further
mention in this connexion.
Otto Frederik Muller was born in 1730 in Copenhagen, where his
father was a musician. He grew up in poverty, was given an opportunity
of studying theology at the university, and then went in for jurisprudence,
but the whole time he had to earn his living as a tutor in aristocratic fami-
lies, on whom his amiable social qualities made a particularly favourable
impression. During his visits to their estates he began to interest himself
in nature, particularly in insects, which he collected and described in a series
'^^mall treatises. As private tutor to a young count he had an opportunity
MODERN BIOLOGY 417
of making a journey through Europe, thereby increasing his knowledge and
widening his connexions. Having returned home, he received an oflicial ap-
pointment, but resigned from the Government service upon contracting a
wealthy marriage, and afterwards lived as a private scholar until his death,
in 1784. His most important work was published posthumously. During his
lifetime he was generally regarded as an amiable and kind-hearted man,
though somewhat vain. Of his immense literary production one or two zo-
ological works have preserved his name for posterity.
As will be realized from the above, O. F. Miiller as a zoologist was es-
sentially autodidactic; he had educated himself by studying the writings of
Linn^us, but he devoted his research work to spheres that Linnaeus and his
pupils had overlooked. In two works, Entomostraca Dania and Hydrachna,
he describes in detail, and extremely well considering the period, two hith-
erto entirely neglected groups of Articulata. Still more remarkable are his
two works on the Infusoria, the last of which, mentioned above, was pub-
lished by Fabricius in 1786. In these works he makes an attempt for the first
time to present a systematic description and classification of the Infusoria,
supplemented with detailed diagnoses of genus and species and illustrated
with accurate and finely drawn pictures. Quite a number of them, especially
the larger Ciliata, he has described so well that they are still recognizable
and their names are still in use today. As was usual in his age, he paid but
little attention to the internal structure of the creatures; in regard to the
origin of the Infusoria, he believes in the spontaneous generation of the lesser
forms, while assuming that the larger and more highly developed forms mul-
tiply by reproduction.
O. F. Miiller's contemporaries and immediate successors made a number
of fresh discoveries in the sphere of the Infusoria, as well as various attempts
to systematize the forms already known. The same period that gave rise to
cell research — the eighteen-thirties — provided also a fresh impetus to mi-
crobiology. In this the pioneer was Christian Gottfried Ehrenberg. Born
in 1795 in the neighbourhood of Leipzig, he studied medicine there and was
afterwards given an opportunity of making a six years' voyage of explora-
tion to the East, whence he brought home important collections. This ex-
pedition having brought him fame, he was invited to accompany Humboldt
on his Asiatic expedition, after which he became professor of medical his-
tory at Berlin and secretary to the Academy of Science there. He died in 1876,
having long given up active participation in scientific developments.
Infusoria as " complefe organistns"
Ehrenberg's great contribution to biology was his work on the Infusoria,
the results of which were published originally in a number of brief essays
and afterwards in the important and splendidly got up work entitled Die
Infusionstiercben als vollkomtnene Organismen, printed in 1838. The result of this
4X8 THE HISTORY OF BIOLOGY
and other works of his was that the number of known Infusoria was consider-
ably increased, and their classification essentially advanced. Anguillulidas
and cercaria, which had hitherto been counted amongst them, were excluded,
the Rotatoria were separated, and the genus- and species-diagnosis pre-
cisely defined, many of them still holding good. The whole of this careful
and praiseworthy work, however, Ehrenberg used in support of an utterly
unprofitable theory; starting from the then prevalent belief in one single
primal type for all animals, he tried to discover in the Infusoria the same
organs as in the higher animals; the vacuoles that are visible in the Ciliata,
and which partly alter shape, were seen in his imagination to possess canal-
shaped outlets and were thus made the basis of an artificially ramified di-
gestive system; he believed he had found sexual organs and eggs in the objects
he investigated — in a word, they were to his mind, as the title of his work
indicated, "complete organisms." That he entirely rejected the belief in the
spontaneous generation of such creatures is self-evident; indeed, this dis-
belief in the spontaneous-generation hypothesis may have been firmly rooted
in him before he began to study the Infusoria and perhaps contributed in
some degree towards inducing in him these efi^orts at finding in them as
complete a form as possible. His theory won many adherents among his
contemporaries — it was embraced, inter alia, by Owen in his earlier works;
when eventually it was exploded, Ehrenberg, after spending some years in
vainly defending his cause, withdrew entirely from all research work.
The scientist who from the outset came forward as a decided opponent
of Ehrenberg and who rapidly won a victory for his views was Felix Du-
jARDiN (i8oi-6i), professor first at Toulouse, then at Rennes. In certain of
his works, the last and most comprehensive of which is dated 1841, he laid
the foundations of a new conception of the Infusoria. He achieved this first
of all by incorporating with them a category of still lower organisms, which
he made the subject of special investigation, namely the Rhizopoda. These,
which include types without any external organs, and indeed without any
definite external bodily form, offered the best possible proof against the
lowest animals' acceptance as "complete organisms." Dujardin found that
both these and the higher Infusoria consist of a homogenous mass, which
possesses the power of absorbing nourishment, contracting and moving, and
reacting to external impressions. This mass he called "sarcode,"a name that
\yas at one time used, especially in France, to denote that fundamental sub-
stance of which living creatures in general are built up, until it was sup-
planted by the word "protoplasm." In the sarcode Dujardin found vacuoles
and granules, but no permanent organs, and the cilia; that cover the body
of the higher Infusoria possess in his view no affinity with the hairy forma-
tions of the higher animals. In all this Dujardin stood undeniably on surer
ground than Ehrenberg; on the other hand, he failed to elucidate the
MODERN BIOLOGY 419
Infusoria's character of simple ceils; their nuclei, which Ehrenberg believed
to be sexual organs, Dujardin was unable to explain more exactly, but
considered them to be simply concretions in the sarcode.
Siebold on Infusoria
It was reserved for Siebold to put the Infusoria in their right place. As early
as his Comparative Anatomy of 1845, which has been referred to above, he
combines Infusoria and Rhizopoda under the common term Protozoa and
characterizes them as ' ' Tiere, in ivelchen die verschiedenen Systeme der Organe nicht
scbarf aiisgeschieden sind, und dercn unregelmdssige Form tmd einjache Organisation
sich auj eine Zelle rediixjeren lassen.'' This definition he bases on a careful de-
marcation of the forms included in the group; the Rotatoria are definitely
separated as being more highly organized, and a number of multicellular,
but primitive life-forms, which produce chlorophyll — Closterines, Vol-
vocines — are transferred to the vegetable kingdom. It is pointed out in
connexion therewith that cilia- and flagella-movements can also exist in
the vegetable kingdom, while on the other hand a special free mobility of
a higher type is attributable to the Protozoa. The various organic systems
that Ehrenberg ascribed to the Infusoria are examined and rejected; there
thus remains the simple cell, provided with nucleus and vacuoles, which is
hereby proved to be capable of sustaining a free and independent existence,
being reproduced by division without any special sexual organs.
In a paper published some years later Siebold further examines the exist-
ence of and the relation between single-celled animals and plants, being sup-
ported in his views particularly by a work by Nageli, which had then been
recently published, on unicellular algx, in which these organisms had been
thoroughly characterized and described. As a result of these works micro-
biology was directed on the right way and during the next-epoch came to
exercise a great influence on the development of biology in general. In the
Protozoa, "the primary animals," had been found a category of living
creatures from which, as the name implies, the other higher organisms
could be derived, besides which the cell idea was hereby made to cover an
entirely new area; it was possible to see in the cell not only the basic element
in the structure of the organisms, but also a true elementary organism capable
of leading an independent life or, as a transition to the higher cell-structures,
of forming colonies of similar elements, such as the Volvox referred to above.
Space forbids our going further into the maze of works which were now de-
voted to the single-celled animals and plants. During the succeeding period
in particular, research work in this field went on apace without interruption.
Of the works on the Protozoa that appeared during this era may be mentioned
those of Friedrich Stein (1818-85), professor at Prague, on the Infusoria,
which form the basis for all later research on the subject.
There still remains to deal with in this connexion one more group of
430 THE HISTORY OF BIOLOGY
single-celled organisms — namely, the bacteria or Schizomycetes, which,
as is well known, have since been found to play a most vital part in human
life and which have accordingly been investigated ever since the middle of
the last century as a special field of research. In connexion with this branch
of research there have existed a number of theoretical problems of the greatest
significance; the problems of spontaneous generation, fermenting processes,
and the origin of various diseases. The problems of the causes of disease
can of course be dealt with only cursorily here; they have of old formed a
science of their own — • pathology. The questions of spontaneous generation
and the fermenting process, on the other hand, have possessed immense
theoretical interest and on the decisive occasion mentioned below their
treatment has happened to coincide. We may therefore suitably begin our
review of the history of bacteriology with a glance at these two questions.
Bacteriology and spontaneous generation
The belief in spontaneous generation has been mentioned on various oc-
casions in the foregoing: how the earlier naturalists generally believed that
the lower animals, especially such as appeared suddenly and possessed more
or less the characters of parasites or vermin, could arise through some kind
of transmutation process in lifeless matter; Aristotle believed that fleas
and mosquitoes originated in putrefying matter — a belief with which even
Harvey at least partially associated himself, while van Helmont had seen
rats arise out of bran and old rags. In the seventeenth century Francesco
Redi (16x6-98), court physician and academician in Florence, proved that
worms in rotting meat arise, not in consequence of the putrefaction, but out
of eggs laid by flies on the meat; if the latter is protected with thin cloth, no
worms arise in it, in spite of the putrefaction. On the other hand, Redi be-
lieved in the spontaneous generation of intestinal worms and gall-flies. For
theoretical reasons Swammerdam denied spontaneous generation; the doc-
trine of preformation that he founded actually precluded any belief in this
kind of propagation and during the greater part of the eighteenth century
held it in discredit. Nevertheless Buffon, as we have seen in a previous sec-
tion, believed in a spontaneous generation through minute life-units scattered
throughout the universe and was supported in his belief by his friend the
English microscopist Needham; and Lamarck associated himself with this
view. On the other hand, Spallanzani, the preformationist, strongly op-
posed the doctrine of spontaneous generation and sought to prove that by
boiling organic elements in air-tight vessels it was possible to prevent living
creatures from arising in them. This theory was put to practical use some
decades later by a French chef, Appert, who invented the still commonly
practised hermetical inspissation of food. A French physicist, however,
found out that the air in the preserving jars lacks oxygen — this element is
really consumed by oxidation processes in the contents — and concluded
MODERNBIOLOGY 43 1
therefrom that the sterility was due to lack of oxygen. At the beginning of
the nineteenth century the belief in spontaneous generation received fresh
impetus, not least as a result of the victory of Wolff's epigenesis theory;
Rudolphi believed in the spontaneous generation of tapeworms, and the
entomologists held the same belief in regard to parasites from the insect
world. It was chiefly, however, the increased knowledge of the Infusoria
that strengthened the belief in spontaneous generation; Ehrenberg's protests
died away unheard when the exaggerations in his description of the organi-
zation of these animals had been made manifest.
Chemists on fermeriting
The idea of spontaneous generation received fresh support in the increased
knowledge of the fermenting process. Both Lavoisier and Berzelius had
studied the fermentation of alcohol and had sought to ascertain the process
of sugar-disintegration in alcohol and carbonic acid; the yeast that floats up
when it is brewed was believed to consist of albuminous elements, which
were separated upon the decomposition of the malt. This conception of
fermentation as a purely chemical process found support in the discovery of a
substance existing in malt that, when added to a solution of starch, converts
the starch into sugar. This substance was called "diastase," and similar
substances, "ferments" as they were called, were soon discovered in other
quarters: in saliva and intestinal fluids in man and animals, as also in many
places in the vegetable kingdom. Chemists believed that they now had in
their hands the substances that produce fermentation and similar processes,
and these chemical changes, in the course of which albuminous compounds
were formed as a by-product, also gave a clear indication as to the direction
in which the spontaneous generation of minute creatures might be looked
for; fermentation was in fact a part of the process of spontaneous generation.
Then there appeared, in 1836, the Frenchman Charles Cagniard de
Latour (1777-1859), an engineer by profession, with the assertion that the
yeast really consists of minute organisms and that it is their activity that
causes the fermentation. Shortly after this the question was taken up by
Schwann, who tried to demonstrate by experiment that putrefaction and
fermentation are processes which are not due to the oxygen in the air, but
rather to a special element that exists in the air and is destroyed by heating;
he boiled special easily decomposable organic substances and then brought
them into contact with air that had first passed through a red-hot pipe,
whereupon no chemical change took place, while there was a change if
ordinary air was allowed access to them. The leading chemists — Berzelius,
Wohler, Liebig — regarded these theories as a chimera, and they won the
day all the more easily because Schwann's experiments were, from the tech-
nical point of view, rather poor; other investigators repeated them and ob-
tained results quite different from those published by Schwann. Thus matters
432. THE HISTORY OF BIOLOGY
Stood when a scientist arose who as the result of unchallengeable experiments
clinched the matter once and for all and thereby entirely changed the course
of microbiology.
Louis Pasteur was born in i8ix at Dole, a town in the ancient province
of Franche-Comte. His parents belonged to the industrial class; his father
had served as a non-commissioned officer under Napoleon and after his de-
mobilization had entered the tanning trade, moving his business from one
place to another. Young Louis attended the country school and afterwards
studied, under circumstances of privation and with numerous interruptions,
in Paris. He was mostly interested in the natural sciences and the teaching
profession was his aim in life. He became a teacher at the gymnasium at
Strassburg and married a daughter of the rector of the school. There he
carried out his first chemical work, which procured his removal to the then
newly-founded University of Lille, where he became professor of chemistry
in 1855. Four years later he was called to Paris, to the Ecole Normale. At
the same time he was carrying on his investigations into the fermentation
process, which at once brought him world-wide fame, and after that, success
and honours came to him rapidly. It was given to him in a specially high
degree to make the results of his investigations of practical use to mankind,
whether it was a question of inventing a method of preserving food, combat-
ing the diseases of domestic animals, or treating rabies, which had hitherto
been considered incurable. This last-mentioned discovery made him particu-
larly popular: through an international fund there was founded in 1889 an
institute for the purpose of investigating those fields of research to which
he had devoted himself, which bore his name and to the management of
which he afterwards devoted all his energies. Previously pupils had already
flocked to his laboratory from all parts of the world; many of them have
themselves won great fame. Nevertheless, this brilliant career had certainly
not been entirely free from shadows. In his political views Pasteur was a
conservative and a warm partisan of the French Empire, and moreover a
strictly faithful Catholic. This caused him a good deal of unpleasantness,
owing to radical opposition, producing an atmosphere that is even reflected
in the scientific polemics waged against him. Ultimately, however, these
hostile voices were silenced, and the more easily as he never meddled in
political questions. But instead he suff'ered in his old age from increasing ill-
health; as early as in 1868 he had had a paralytic stroke which impaired the
use of one of his arms, but which did not succeed in preventmg him from
continuing his activities with as great success as before; gradually, however,
his powers declined and he passed peacefully away in 1895. His name lives
in his work and in the "Pasteur Institutes" which have been established in
all civilized countries; he is without doubt one of the greatest scientists of
his century.
MODERN BIOLOGY 433
Pasteur denies spontaneous generation
Pasteur began his researches in the purely chemical sphere; he investigated
organic acids, chiefly the isomeric, and he obtained valuable results in con-
nexion therewith. Thence he was led to study the question of the molecular
structure of sugar and the manner in which this substance is converted into
different isomeric alcohols and acids — in other words, into the process of
fermentation. His first experiments in this field were concerned with the
formation of lactic acid; he found on the surface of sour milk minute greyish
spots, which he examined microscopically and experimentally. Under the
microscope they appeared as a mass of minute globular formations, smaller
in size than those of which ordinary yeast is composed; when placed in a
saccharine solution they at once disintegrated it into lactic acid. Without
at the moment drawing any conclusion as to their origin, he maintains that
all fermentation is caused by similar minute organisms. If we plant out such
organisms of a definite type in a saccharine solution, wt get a definite form
of fermentation: alcoholic fermentation, br.tyric-acid formation; if, on the
other hand, a suitable saccharine solution is allowed to stand by itself, there
are set up a number of simultaneously disintegrating processes induced by
different organisms operating at the same time.
It goes without saying that the chemists of the old school did not feel
that they had much to gain from these new discoveries. They at once found
a keen supporter in Felix Archimede Pouchet (1800-71), professor at Rouen,
who was reputed both as a botanist and as a zoologist. In a series of investi-
gations he tried to prove that the micro-organisms arising upon fermentation
and putrefaction are spontaneously generated, and this owing to the very
fact of those chemical changes; the fermentation forms the initial stage of
the process whereby living creatures arise from the decomposition of existing
organic substance. In the view of such a theory Pasteur's fermentation experi-
ments were, of course, pure irrational nonsense, and thus began a lengthy
controversy betw^een these two experimental scientists, in which the scientific
world and eventually the enlightened public became keenly interested.
Controversy between Pasteur and Pouchet
To Pasteur the specific character of the different fermentative organisms was
in itself a proof that they are not the product of chemical change, but actual
species of living beings , which come into existence through the multiplication
of existing individuals. But whence, then, come all those different organisms
which immediately populate saccharine solutions that are allowed to stand,
and food that is kept too long? Schwann had derived putrefaction from the
air, and Pasteur endeavoured to prove this by experiment; he filtered air by
sucking it in through a tube filled with cotton-wool, thereby obtaining a
collection of dust particles, which were transferred with great care to a
retort filled beforehand with boiled and cooled saccharine solution; the neck
43 4 THEHISTORYOFBIOLOGY
of the retort was melted together and after some days there was found in the
fluid an abundant vegetation of micro-organisms. Their origin was thus
proved; these creatures had existed in atmospheric dust in a dried state. On
the other hand, a saccharine solution boiled in a retort the neck of which
was melted together in the course of boiling could be preserved for any
length of time without changing. Pouchet tried by various means to dis-
credit these experiments; he tried to prove that the organisms cannot stand
being dried, that they do not exist scattered in the air as Pasteur declares,
that milk becomes sour in spite of being boiled. Space forbids our following
this dispute in all its phases — how both parties collected air on high
mountains with a view to proving their arguments on that evidence; how
Pasteur found that certain organisms can endure heating up to the boiling
point of water without perishing (which explains how it is that boiled
milk turns sour); how he thought out a whole series of ingenious apparatus
to prove his statement that the fermenting organisms always originate in
the outer air and that the boiling of the experimental fluids and the heating
of the air which comes into contact with them, infallibly exclude the exist-
ence of organic life in them. The two antagonists were allowed to carry out
their experiments before the French Academy of Science, and Pasteur at once
succeeded in convincing some of its foremost members — Milne-Edwards
and Claude Bernard, and the chemist Chevreul. Pouchet likewise had his
supporters, and especially among the scientifically educated and half-educated
public he gained many adherents who regarded spontaneous generation as a
"philosophical necessity," indispensable for a natural-scientific explanation
of the origin of life, which Pasteur, faithful Catholic as he was, naturally felt
himself compelled to explain dogmatically. Thus argument opposed argu-
ment, and party faced party. In these circumstances the solution of the prob-
lem would never have become possible had not Pasteur been able to put his
ideas into practice on a large scale. During the succeeding years he invented
his well-known methods of preserving milk by "Pasteurizing" — that is,
by heating — of improving the manufacture of wine and beer by controlling
the conditions of fermentation, of securing immunity from the silk-worm
disease and chicken cholera by eliminating the micro-organisms that produce
them. These discoveries, however, belong to the next period, as also the de-
velopment and perfecting of bacteriology and fermentation research achieved
by other investigators — Koch, Hansen, and many others. Finally Pasteur's
views on the origin of the micro-organisms received splendid practical con-
firmation as a result of the development of modern medicine: antiseptics and
aseptics in surgery, disinfection, and the treatment of infectious disease.
Owing to these facts, which found fresh confirmation daily, spontaneous
generation has entirely ceased to exist as a possibility to be reckoned with in
modern biology, nor does it come into serious question when we have to
MODERNBIOLOGY 43 5
explain actual phenomena. That its theoretical possibility nevertheless still
continues to be keenly discussed is due to modern natural-philosophical
speculation — a subject that will be dealt with in a later chapter.
4. Botany
Plant classification after Linnaus
A SURVEY of the most important data in the history of botany, particularly
of plant classification, up to the period covered by our chapter heading, is
necessary if we are to obtain a universal view of the development of biology
during the period now being described. In order to obtain this view we must
first of all return to the days of Linnasus. It will be remembered that Linnaeus
set up, in the first place, an artificial system based exclusively on the structure
of the various parts of flowers and especially intended to be used for practical
examination purposes, and, in the second place, a natural system, based on
the common forms of plants — a system which he worked at throughout
his life, without, however, being able to find a satisfactory conclusion to it.
His immediate successors paid but little attention to this latter legacy from
him, although it really offers immense possibilities for development; they
contented themselves rather with examining as many new plants as possible
according to the sexual system. During this time, however, there were
published a considerable Tiumber of sound systematic works; the study of
cryptogams in particular made rapid progress, but nothing was contributed
to the development of the system itself until twenty years after Linnaus 's
days, when Jussieu's systematic work. Genera -plantarmn, was published.
Antoine Laurent de Jussieu came of a family that had already given
France two eminent botanists, principally in the spirit of Tournefort; es-
pecially Bernard de Jussieu, uncle of the above-named, who had made con-
siderable additions to a natural system, without, however, succeeding in
completing it. A. de Jussieu was born at Lyons in 1748, studied medicine in
Paris, and became professor at the Jardin des Plantes, after which he held
some other botanical and medical posts and finally became professor of phar-
macy in the faculty of medicine. He was active for many years and attained a
great age; he died in 1836. In his principal work, mentioned above, Jussieu
has set up a complete natural vegetable system, the first of its kind. Like Ray,
he makes the cotyledons the chief basis of classification for the vegetable king-
dom; in his view this is justified, because the plant arises out of the seed, and
the latter's most vital part is the cotyledon — it is compared with the heart
of animals. Consequently plants are divided into: acotyledons, monocoty-
ledons, and dicotyledons, a system of classification that, thanks to him, has
become permanent. These three main divisions are then divided into orders.
43 6 THE HISTORY OF BIOLOGY
the names and demarcations of which are taken partly from Linn^us's plan
for a natural system and partly from the preliminary work of Antoine's
uncle Bernard. Some of these orders are, in fact, quite natural, while others
are extremely ill arranged, especially in the acotyledon group, in which are
included not only Linn^eus's cryptogams, but also the naiads; thus, he in-
cludes the ferns among the Cycas, which is placed between the Polypodia
and the Equiseta. Jussieu has formed his genera mainly after Tournefort,
while his definition of species is reminiscent of Ray rather than of Linnasus.
Jussieu, therefore, was not a very original observer, but his service to science
lies in the fact that he really worked out a natural system, which he set up
in determined opposition to Linn^us's sexual system, and which has, in
fact, been the starting-point for all subsequent systematic improvement.
A far more important observer was Robert Brown (1773-1858), who
has been mentioned before as the discoverer of the cell-nucleus. The son of a
Scottish clergyman, he studied medicine in Edinburgh and became an army
surgeon, but at the same time he applied himself to botany and was appointed
botanist to an expedition to Australia, which was led by a Captain Flinders.
Brown stayed four years on that continent and brought home large collec-
tions; on his return he was made librarian to the Linnean Society and cura-
tor at the British Museum. In this position he enjoyed the reputation of
being one of the finest botanists of his time; he never published any very
important work, however, and his papers were, curiously enough, collected
and published in a German translation, done by Nees von Esenbeck, with
whom he had but little in common from a scientific point of view. Nor did
he work out any system of his own; in his works heuses sometimes Linnasus's,
sometimes, and more often, Jussieu's natural system. His service to science
lies in the care and keen-sightedness with which he works out and analyses
the various orders, or families, as he more frequently terms them:^ his studies
of the Compositic, Asclepiadace^, and many other families have been men-
tioned as models for the research work of the succeeding age and have con-
tributed much towards finding a place for the natural system in the scientific
mind. Brown was, moreover, an eminent plant-geographer and made a special
study of the distribution of the different families in different climates; in this
respect his earlier work on the flora of Australia was unrivalled and attracted
the attention of Humboldt, who highly commended it.
As one of the foremost pioneers of botany should also be mentioned
AuGusTiN Pyrame de Candolle. He was born in 1778 at Geneva, where his
family had for generations enjoyed a great reputation. At an early age he
began to study the natural sciences, which at that time — the age of Bonnet
^ The term family" as an expression for the natural groups in the vegetable kingdom ap-
parently comes from the French botanist Michel Adanson (i 717-1 806), whose attempt, influ-
enced by Buffon, to form a natural system for the vegetable kingdom was somewhat of a failure.
MODERN BIOLOGY 437
and Saussure — stood in high favour in his native town. After preliminary-
studies there, he betook himself to Paris in order to continue his education
as a botanist. In the company of Lamarck, Cuvier, and Geoffroy he spent ten
years there, during which his reputation increased year by year and public
commissions were entrusted to him; amongst other things he was sent, with
the financial assistance of the State, on scientific expeditions in different parts
of France; Lamarck handed over to him the editing of his French flora and
he was finally elected professor at Montpellier. In 1816, however, he returned
to Geneva, which during the Revolution had become incorporated with
France, but after the fall of Napoleon was again united to Switzerland. He
then lived in his native town as professor of botany and member of the high
council, honoured and respected until his death, in 1841.
De Candolle on -plant morphology and physiology
De Candolle mastered the whole field of botany better than anyone else in
his time; he was at once systematist, morphologist, and physiologist. He
started a gigantic work, Prodromus systematis naturalis regni vegetabilis, which
was to describe all known plants, but which for obvious reasons was never
completed in his lifetime; his son and many others worked at it after his death.
The principles on which he classified the vegetable kingdom he laid down
in a work published in 18 13 entitled Theorie elementaire de la botanique, which
he revised several times and which is without doubt his finest work, worthy
'to be associated with, and at the same time representing a great advance on
Linnasus's Philosophica hotanica, which doubtless gave him the idea. It starts
with a general scientific theory, according to which nature is controlled by
four great forces : attraction, and affinity, which are the basis of physical and
chemical phenomena, the life-force, which is common to ail living creatures,
and sensibility, which is the characteristic of animal life as opposed to vege-
table life. Each of these four forces has its own science: physics, chemistry,
physiology, and psychology. It is thus a markedly vitalistic conception,
which is still more emphasized in such an assertion as that "the life-force
annuls or modifies, as necessity dictates, the ordinary laws of matter." De
Candolle is by no means a fantastic natural philosopher, however; on the
contrary, he has the same sober and critical conception of natural phenomena
as Cuvier, whose correlation theory he applies to the vegetable kingdom.
He maintains that the two most vital organic systems in the plants, the vege-
tative and the sexual, are dependent upon one another; a plant with highly
developed fertilizing organs cannot possess primitive vegetative organs, and
vice versa; therefore a natural system, set up with comprehensive regard to
the entire reproductive organization, should at once conform to such a
system set up with a view to the vegetative organs, and this, indeed, he
proves by means of examples. In connexion herewith he maintains, under ac-
knowledgment to Linnasus, that there is throughout the vegetable kingdom
43 8 THE HISTORY OF BIOLOGY
a universal symmetry, a standard of organization, which is modified in the
individual by the same organ's being capable of serving different purposes
and the other organs' undergoing corresponding changes. Among these
changes he includes especially stunted growth, degeneration, and accretion;
in his view a flower with free petals is higher than one with accrete petals —
a principle which he applied in his system, though it failed to gain the ac-
ceptance of posterity. For the rest, he introduced into his system reforms of
lasting value; the difference between vascular and cellular plants was estab-
lished by him, as also the contrast between the bole-plants and the higher
plants. Further, his classification of the dicotyledons has been largely ac-
cepted by subsequent botanists. Otherwise, de Candolle strongly repudiates
Lamarck's theory of one single evolutional chain in the organisms, instead
associating himself with Linnasus's idea of the natural system's likeness to
a map; in fact, his idea of species is not unlike the Linnasan: according to
de Candolle, a species is "the sum total of all the individuals which mutually
resemble one another more than they resemble others, which are capable by
mutual fertilization of producing fertile individuals, and which are multi-
plied by generation, so that it is possible by analogy to assume that they
have originally sprung from a single individual." Varieties arise, he con-
siders, partly through the influence of local conditions of life and partly
through hybridization; moreover, there are in certain quarters varieties
which must be regarded as constant, like the species, and which should be
distinguished from the accidental local varieties. The genus is defined as a
collection of species with a striking mutual resemblance in regard to all
organs; families and higher categories are given similar definitions.
Among de Candolle 's other works may be mentioned his Organography,
an account of the organic systems of plants, and his Physiologie vegetale, a
work of great merit for its time, based on a thorough knowledge of the vital
conditions of plants and of chemical and physical processes belonging thereto.
We must, however, pass over these works here; as a matter of fact, it is
mostly as a reformer and theoretician in the sphere of classification that
de Candolle has made his best contribution to the development of biology.
End lie her' s sysfem
The development of plant classification received further impetus through
Stephan Ladislaus Endlicher (1805-49). Bo"*^ of wealthy parents at Press-
burg, he studied first of all theology, but afterwards devoted himself both
to botany and oriental languages. He became professor of botany and head
of the botanical gardens in Vienna, acquiring fame for the splendid initiative
he took in furthering the development of natural science in Austria, gener-
ously contributing towards that end out of his own private purse. He pre-
sented his herbarium to the State and published a botanical journal at his
own expense. At the same time he made a name for himself as an expert in
MODERNBIOLOGY 43 9
the Chinese language. As a teacher he won the affection of his pupils. In the
year of revolution, 1848, he took advantage of his popularity to plead the
Government's cause before the rebellious students, but they became em-
bittered against him and drove him out of Vienna. This he felt very deeply
and he died shortly afterwards, some say by his own hand. His great work
on plant classification is his Genera plantarum, which comprises all the then
known vegetable genera arranged in a natural system; he has given a brief
summary of the subject in his Enchiridion. His service lies not so much in the
new ideas that he produced in the sphere of classification as in the particu-
larly clear, concise, and complete characterization and demarcation of fami-
lies and genera which he created and which made his work the basis for all
later plant classification.
Hedwig on mosses
By the side of this generally systematic and morphological research there
developed a keen interest in the specialized study of particular botanical
subjects, the hitherto neglected lower plants offering, of course, an especially
attractive field of study. As a pioneer in this sphere Johann Hedwig (1730-
99) is worthy of mention. He was born in Hungary, but he worked mostly in
Leipzig, first as a physician and then as professor of botany. He applied him-
self to the study of the multiplication of the cryptogams, making careful
observations of the propagation and germination of the spores. It was
chiefly. the mosses, however, that occupied his attention, and in this field
he was a pioneer; he divided the large and unwieldy genera that Linnasus
had created into a number of well-characterized genera, which are in part
still retained, and he found for them a good basis of classification in the shape
and marginal formation of the capsules. Many other naturalists have since
followed in his footsteps, so that muscology is now a thoroughly elaborated
specialized section of botany.
The Algas and the Fungi became subjects of special treatment much later
than the mosses. Of the pioneers of Algx-research we have already described
one of the foremost, Carl Adolf Agardh (Part II, p. u^i). His son, Jacob
Georg (1813-1901), followed in his father's footsteps with credit.
Vries on fungi
In the Fungi as a field of research Sweden has also produced one of the most
eminent names, that of Elias Fries, who has likewise been one of the most
distinguished Swedish botanists since Linnxus. Born in 1794, the son of a
priest in the province of Smaland, Fries devoted himself, even as a boy, to
the study of botany, and of Fungi in particular; when still a youth he became
a lecturer at Lund and in the year 1834 professor at Upsala, where he was
active until 1859, when he became professor emeritus; he died there in 1878.
When Fries first went to Upsala the University was a centre of romantic
reverie and metaphysical speculation; he took up the cudgels with success
440 THE HISTORY OF BIOLOGY
and honour on behalf of the cause of exact research, by no means allowing
his own special sphere to be put in the shade; in speeches and writing he
championed the cause of biology, and his plea was heard far and wide.
Being a brilliant stylist and an eloquent speaker, he was elected a member
of the Swedish Academy, and in his old age he held the position of a recog-
nized patriarch in the sphere of natural science in his own country. Although
no friend of fantastic speculations, he shared his age's idealistic conception
of nature and thus found it easier to gain a hearing for his high aims; in one
of his writings he expressly calls biology a "supernatural" science, for life
is something higher, given from above, and its influences must not be ex-
plained according to the laws of inorganic nature. To his mind, biology
belongs not to the exact sciences, but to the historical; it is more closely
akin to theology than to physics and chemistry.^ In his special research work,
however. Fries is quite exact. His chief productions are his great works on
the Fungi, which have since formed the basis of classification in this class;
he has described quantities of species and given characters to genera and
families that still hold good. Next to these works should be mentioned his
treatises on the lichens; here he had a precursor in his fellow-countryman
Erik Acharius — born in 1757 and mentioned as Linnseus's last pupil,
afterwards provincial physician at Vadstena, died in 18 19 — but it was Fries
who established the lichen system, which was generally accepted until the
eighteen-sixties. Fries also performed a considerable service in producing
his classification of the phanerogams; among other things he maintained,
in opposition to de Candolle, that the Compositie are the highest-developed
of the phanerogams, and in this posterity has shown him to have been right.
His natural system has, with certain modifications, been generally utilized in
Scandinavia.
At this point our description of the different spheres of biology has
brought us up to the period that is characterized by the launching of the
theory of the origin of species. It now remains for us to give a glance at
certain phenomena in the sphere of theoretical speculation that have given
direction to our modern biological views.
^ Botanical Excursions {Botaniska utflykter^ I, pp. 11-13. Curiously enough, Haeckel from
his standpoint arrived at similar conclusions, of which more anon.
CHAPTER IX
POSITIVIST AND MATERIALIST NATURAL PHILOSOPHY
Romanticism and positivism
IN HIS History of the Philosophy of Later Times HofFding declares that there
are two intellectual currents characteristic of the nineteenth century: ro-
manticism and positivism, the former starting from the ideal of thought,
the latter from that which is based on fact. This division is undoubtedly in
accordance with the actual course of events dominating the whole world of
culture; the contrast indicated is discerned no less clearly in the development
of biology. The romantic conception of nature that prevailed at the begin-
ning of the century saw the true reality in an idea, of which the actual life-
forms were merely modifications; they sought therefore for a primary form
or archetype, with which the living forms were compared, as was done, each
in his own way, by Goethe, Geoffroy Saint-Hilaire, and R. Owen, the last-
mentioned still as late as towards the middle of the century. By that time an
entirely different conception of natural phenomena had already appeared,
which fought its way year by year into the general consciousness; although
champions of the old ideal still survived far into the latter half of the century,
nevertheless it may be claimed that the victory of the new conception was
already fully confirmed by the beginning of the sixties. Opposition to the
old concept first came from the social and political spheres, after which it
took in its stride the scientific and literary world. Its original home, there-
fore, was in the two countries in which public life manifested the greatest
mobility — France and England. It was not until later, and then under
different forms, that it appeared in Germany and Scandinavia.
In France there set in during the time of Napoleon, and still more im-
mediately after that era, a violent reaction against those radical ideas that
the enlightenment of the eighteenth century had created and the Revolution
had sought to realize — a reaction especially in the social and political
spheres, less in the scientific, although it certainly had its learned theorists,
as, for instance, the brilliant and fanatical Joseph de Maistre. But neverthe-
less the theories of the period of enlightenment could never be wholly
suppressed; they survived, as did the longing for the political freedom of the
Revolution, and they found support in the natural sciences, which at that
time were passing through a brilliant phase in France, being sustained by men
who worked for the most part undisturbed by any theoretical speculations.
441
442. THE HISTORY OF BIOLOGY
In a previous chapter we have described the most important repre-
sentatives of biology in France during that period; physics, chemistry, and
astronomy at the same time could boast of no less brilliant representatives.
And the results that these sciences achieved were very clearly brought home
to the world in general, owing to their splendid application to practical
life, which was then just beginning and which afterwards represented per-
haps the most striking characteristic feature of the century; it was then that
the influence of steam-power on industry and communications first began
to be realized; it was then that the significance of chemistry in numberless
fields of activity began to make itself felt in the daily life of humanity,
not to speak of the somewhat later application of electrical phenomena in
practical everyday life. As a result of all this the natural sciences began to
influence the public mind more than they had ever done before; mankind ex-
pected them to lead to new and happier times, while theology and philoso-
phy, which had served the oppressors of the people, reaped nothing but
hatred and contempt. Would it not be possible for all forms of human life,
for the whole of human culture,, to be placed under the a^gis of the natural
sciences, to be explained through them and developed in their spirit? This
question was answered by many with an unreserved yes; foremost of these
was Comte, one of the most gifted thinkers that France has ever produced
and certainly the most influential during the past century.
Isidore Auguste Marie Francois Xavier Comte was born in 1798 at
Montpellier. He belonged to an ultra-Catholic and strongly conservative
family and received a strict upbringing in the same spirit. Even when he was
fourteen years old, however, he began to doubt the correctness of the dog-
mas on which he had been brought up, and when, later on, he went to study
at the Ecole Polytechnique in Paris, his oppositional attitude became clearly
defined and was all the more strengthened when the Government disestab-
lished that institution, which was feared as a centre of opposition, before
he had had time to complete his studies there. Thus he passed no examina-
tion, a fact that had a disastrous efi^ect upon his future; nevertheless, in spite
of his parents' opposition, he continued his studies in Paris, steadily im-
proving the substantial knowledge he had already acquired of mathematics,
physics, and chemistry. He soon won a reputation for his genius and eru-
dition — among his patrons and friends he counted men such as Humboldt
and Blainville — yet he was never able to obtain a post in the Government
service, but all his life he had to earn his living by private tutoring, except
for some years when he was employed as an assistant teacher. This may to a
certain extent be explained as due to the peculiar theoretical point of view
he adopted; even in his youth he resolved to devote his life to creating a
general system, which was to deal, along natural-scientific lines, with the
whole of existence, both of nature and of human life, and thus to arrive at
MODERN BIOLOGY 443
a universal knowledge, whereby all the problems of life would be solved,
not only the theoretically scientific, but also, and above all, the social and
political. The first concept of this system was drawn up in the form of private
lectures, to which he succeeded in attracting a large number of listeners;
then during the period 1830-4^ he worked out his famous Cours de philosophie
positive in six large volumes. His method of working was peculiar to him-
self; trusting to his phenomenal memory, he had recourse to no literature
of any kind when at work, and he used to write down the contents of a
whole volume at a time without corrections, after having worked out in his
head the gist of what he was going to write. It is natural that in such cir-
cumstances his work should be full of inaccuracies in matters of fact as well
as of stylistic redundancies. In several cultural circles the influence of the
work has been deep; in view of the important part that biology plays in it,
it is worth giving a summary of the book here, all the more so as it had its
effect on the biological theories of the succeeding era.
Comte s positive philosophy
It is a "positive philosophy" that Comte desires to create; by "philosophy"
he means, as did Aristotk; whom he greatly admired, a knowledge of the
whole of existence; by "positive" he means "/<« 77ieme chose que reel et utile.''
This "real and useful" knowledge he will substitute for the theological
and metaphysical, which he considers to have predominated during previous
epochs. He finds the essence of existence to be in the development that has
always taken, and is still taking, place; in this instance he paved the way
for the explanation of life that has governed human culture since his time.
In contrast to so many later positivist thinkers, however, he does not look
for this development in nature — it is characteristic that geology does not
interest him at all — but in human life. In the history of human thought
three successive phases have followed one another: the theological, in which
it was believed that personal divine powers were the cause of all that hap-
pened; the metaphysical, when for these were substituted impersonal forces;
and the positivist, in which men no longer ruminate over the causes of all
that takes place, but are content to establish facts and determine their
course. The theological stage culminated in the Catholicism of the Middle
Ages, for which Comte, in spite of all, expresses great sympathy. As the
founders of positivism he cites Bacon and Galileo, whose explanation of
nature should, in his opinion, be applied to all phenomena. The middle
phase, the metaphysical, is, in his view, the worst of all; it is the belief of
the idealistic philosophy in spiritual reality, beginning with Descartes and
ending with the romantic philosophy, that is sharply, and for the most part
justly, criticized. Instead, his explanation of nature is the same as Galileo's —
that it is not possible to find out what the forces of nature are, but only how
they operate. In this he is undeniably right; his weakness, on the other hand,
444 THE HISTORY OF BIOLOGY
lies in his mania for formally reducing to simple formulas all phenomena,
even those of the most complex character. He thus forgets Galileo's second
great exhortation: to measure what is measurable and to make measurable
what is not. He believes, for instance, that every being, and especially every
living being, can be studied from two sides, the static and the dynamic —
that is to say, as potentially active and as actually active. Thus, biology has
a static side, anatomy, and a dynamic side, physiology, and other sciences in
like manner. Conite himself declares that he borrowed this division into
static and dynamic from Blainville; it again occurs in Haeckel's Generelle
Morpbologie. According to Comte, all science should be classified after the
method employed by botanists and zoologists; by this method we get six
separate branches of science: mathematics, astronomy, physics, chemistry,
biology, and social physics, or, in a single word, invented by Comte and
now generally accepted, sociology. Each of these sciences is based on all
the previous ones in the series and cannot be mastered without a knowledge
of them. The biological section is, of course, the one that affords chief in-
terest to the present history.
His biological ideas
As he repeatedly asserts, Comte's biological speculations are most closely
associated with those of Blainville, but are, of course, entirely outside the
scope of the control which the theories of that distinguished zoologist
exercised in his special research-work. Blainville's view that life consists of
"composition et decomposition ' is thus embraced by Comte, who with its
support rejects both Stahl's vitalism and Boerhaave's mechanism. On the
other hand, he accepts Bichat's tissue theory, strongly supporting the idea
of structure's being the essential factor in the living organism; Bichat's
"organic and animal life" is also adopted by Comte. Life itself he defines as
"the relation between organism and environment." It can be studied, as to
both its static and its dynamic side, after three different methods: observa-
tion, experiment, and comparison. Observation is the fundamental method
and should be carried out with all available technical appliances. Experiment,
on the other hand, is condemned, especially vivisection, which disturbs the
relation between the organism and its natural environment and thus merely
creates abnormal states and, moreover, leads to cruelty — Comte does not
mention Magendie's name, but obviously refers to him. — The finest biologi-
cal method is the comparative, which is applicable, on the one hand, to dif-
ferent parts and stages of development in the same individual, and, on the
other hand, to different life-forms. The latter type of comparison should be
concerned with both organs and tissues; Comte assumes a primal tissue from
which all other forms of tissue and organ can be derived, but he rejects the
cell theories that were just then making their appearance. This derivation,
however, turns out to be as idealistic as Cuvier's comparative anatomy;
MODERN BIOLOGY 445
Comte, indeed, maintains with the latter that the species are invariable, "for
the idea of species would inevitably cease to represent an exact scientific
definition if we were to allow an unlimited modification of different species,
the one in the other." This opposition of Comte's to the theory of the origin
of species was undoubtedly the cause of Haeckel's refusing to acknowledge
him as a precursor in respect of monism, which he nevertheless was far more
than any of those whom Haeckel enumerates.
The details of Comte's biological speculations are, of course, of interest
only from the point of view of curiosity. He associates himself with Blain-
ville's animal system, with its exclusive reference to external characteristics;
he himself lays down three main types for the entire animal kingdom:
Osteozoa, Entomozoa, and Malacozoa — that is, Vertebrata, Articulata,
and Mollusca — a classification the clumsiness of which scarcely needs
pointing out; indeed, the Mollusca group in particular, a reversion to Lin-
naeus's Vermes, was at the time utterly absurd. Still worse, however, is
Comte's attempt to analyse "the intellectual and moral cerebral functions,"
for here he becomes infatuated with Gall's phrenology. It is only natural
from his point of view that he should reject Descartes's theory of the parallel
existence of the soul and the body, and the other "metaphysicians" likewise
offer many points of attack. In his criticism of the earlier psychology, then,
Comte has shown very keen observation, but the psychology that he him-
self created is all the more lacking in criticism. He denies the possibility of
psychical self-observation, for one cannot divide oneself into two parts for
the one part to observe the other, and besides one cannot in this way find out
the mental life of the animals, which is the vital preliminary stage to that
of man. True psychology should, according to Comte, be based on Gall's
theory of intellectual and moral areas in the brain, which is the beginning of
an entirely new psychology. Here modern psycho-physical research has pro-
ceeded along a line which Comte never dreamt of and which led to the
complete acceptance of that idea of "self-observation" which he despised.
His sociology
The last three sections of Comte's work deal with sociology, the doctrine
of social statics (that is, organization), and dynamics (that is, progress).
Eventually these problems entirely usurped the place of natural science in his
life's work. It is true that he produced ideas of value in this sphere too: the
actual principle of studying social life from, so to speak, a biological point
of view has indeed won adherents in modern times, and a number of items
in his program, as, for instance, mixed schools, have actually been adopted.
But on the whole his social theory is only a curiosity. This is due mostly to
the strange development that he himself underwent. While still a young
man he had for a year been a lunatic, but he afterwards recovered his mental
balance. When he had concluded his great book, he added to it a general
446 THE HISTORY OF BIOLOGY
introduction which led to his being accused of megalomania and persecutory
paranoia, and after that period he became engrossed in ever stranger social
Utopias; he founded a new religion, "a Catholicism without Christianity,"
as Huxley called it, with catechism, a calendar of saints, comprising great
men to whom prayers should be addressed — beginning with Moses and end-
ing with Bichat and Gall — and a ritual of divine service. His scientifically
educated friends deserted him, and only a small group of a less intelligent type
gathered round him at his death, in 1857.
The influences that Comte exercised upon the conception of life held by
subsequent generations is not easy to estimate. All that in modern times has
gone under the name of positivism, monism, utilitarianism, and various
other isms has either directly or, at any rate, intermediately been influenced
by his doctrines. In conscious opposition to the ideal unity, in which roman-
ticism saw the connexion of existence, he took evolution to be the connecting
force in life. He certainly did not view biological evolution in the same light
as modern biology — if he had, he would not have rejected Lamarck — but
he observed with all the keener vision the evolution that is taking place in
human culture. He was thus able to do justice to the various stages of his-
tory — a thing which the period of enlightenment of the eighteenth century
was unable to do — while, on the other hand, he could point to a goal in
the future towards which to strive. And this belief in the evolution of man-
kind was, as we shall find later on, a precondition before the theory of evolu-
tion in nature could gain a hearing; Comte therefore paved the way for the
doctrine of the origin of species more than most others did. And though his
own biological concept was deficient, it has nevertheless had its influence;
we have already pointed out traces of it in Haeckel, and these could probably
be supplemented; even in later times there has been a corresponding tendency
reminiscent of such characteristics as a preference for comparative investi-
gation and a dislike for experiment. Comte's name has, in fact, a definite
place in the history of biology.
English -positivism
The other representatives of positivism in France — Comte's pupils — de-
voted themselves principally to social and general cultural problems and
may be passed over here, however deep their influence may have been on the
general conception of life, both inside and outside their own country. The
same is to a certain extent true of the precursors of the same realistic trend
of thought in England. That country was indeed the cradle of eighteenth-
century enlightenment, and the ideas of the era of enlightenment never quite
died out there, even in the days of romanticism. These ideas took rather
the form of strivings after practical social reforms, as in Jeremy Bentham
and James Mill, who are named as the founders of utilitarianism, a general
philosophy of life with a social aim, based upon the highest possible happiness
MODERN BIOLOGY 447
for the greatest possible number of people, a happiness that would infallibly
be attained through the activities of the individual being as little as possible
restricted by the various organs of the community. Here, then, we find the
same belief in evolution as in Comte, although in a more practical form. The
foremost supporter of these ideas, however, is John Stuart Mill (1806-73),
son of the above-mentioned James Mill. He was taught by his father and
never studied at a university, but as a young man entered the Civil Service.
For a time he was a friend of Comte and was influenced by him. Among his
works should be first mentioned his System of Logic, an analysis of the laws of
thought, which had a great influence on the generation that felt the first
effects of Darwin's theory; Haeckel especially cites its doctrines frequently.
Mill derives all knowledge from experience, and this in its turn from sense-
impressions; of special interest is his analysis of the different ideas of the
natural-scientific systems, particularly the idea of species ; he considers a well-
defined species to be a reality, not merely a conventional term, but, on the
other hand, he maintains that the species should be based on characters and
not on any imaginary ideal type. The closer study of these extremely detailed
analyses of ideas is, however, more a concern of philosophy than of the his-
tory of biology. For the rest. Mill was active both in theory and in practice
as a liberal social politician and as such possessed a wide influence.
Dotvnfall of romanticism in Germany
The advent of the realistic conception of life took an entirely different turn
in Germany. It will have been seen from the foregoing how education in
that country had for half a century been entirely dominated by the romantic
philosophy, with the result that even natural science came to a great extent
under the influence of its modes of thought. The Schellingian polarity-theory
certainly had very soon to give ground, but in the world of speculation the
Hegelian philosophy, with its dialectical method and its contempt for all
empirical research, prevailed all the longer. But after the death of the master
the school was divided against itself, and many of its members developed
their doctrines along distinctly radical lines, as, for instance, Karl Marx,
the famous founder of socialism, and Ludwig Feuerbach (1804-72.), whose
views closely approached the positivism of Comte and who was otherwise
a thinker mostly engaged in problems of religious philosophy, and also
D. F. Strauss, the well-known Bible-commentator. Other philosophers re-
mained on the old ground, while still a number, including some of the most
keen-sighted, returned to Kant's critical studies. Whereas, then, the roman-
tic philosophy was being internally disrupted, the natural sciences made the
splendid advance that has been described in the foregoing. It is no wonder,
therefore, that natural-science students took courage; the results of philoso-
phy had resolved themselves into vain squabbles; why not, then, let scientific
research be self-sufficient and solve the riddle of existence on its own account?
448 THE HISTORY OF BIOLOGY
Good progress had already been made when science was backed by such
sound victories as Mayer's law of energy and Wohler's organic syntheses.
German materialism
It was in these circumstances that the new realistic natural philosophy arose,
whose different ideas have occupied the attention of so many thinkers and
writers up to the present day, and which has gone under so many different
names, such as positivism, materialism, monism, agnosticism, and other
isms. Its main characteristic has been the endeavour to build up, on an exact
natural-scientific basis, an explanation of the whole 0/ existence — that is, to
base on the limited results of research an explanation of the illimitable, by
means of weights and measures to explain the immeasurable and imponder-
able. These natural explanations might have been fully justified as expressions
of a personal view of life, if their originators had clearly realized the differ-
ence between facts and hypotheses, between manifestations that are actually
capable of being observed and turned to practical use, and theoretical con-
structions of such as are inaccessible to any observation whatsoever. This clear
thinking, however, has unfortunately been somewhat rare; far more common
has been the tendency to work up explanations of nature and then insist upon
having them regarded as the results of natural-scientific research — a weak-
ness that has often been apparent in men who in their own special sphere
have been keen and conscientious observers. From the outset this temptation
was no doubt due to the influence of romantic philosophy, which had con-
fidently proclaimed the infallibility of its absolute natural explanations.
Another factor, especially as regards the more popular scientific literature,
was the rivalry with the ecclesiastical tenets, which maintained the absolute
truth of the words of Scripture, even in questions of natural science. And,
finally, there were in Germany the political points of view to be reckoned
with; the ruling powers did their utmost to preserve the old belief in au-
thority, which was considered to conduce to obedience to government; the
opponents of this belief were consequently on the side of natural science.
The contrast was still further sharpened by the revolutionary outbreak of
1848 and was by no means softened by the stern measures which the Govern-
ments adopted after their victory, in order to maintain their authority. Con-
siderable light is thrown upon these conditions by the so-called materialist
dispute in the beginning of the eighteen-fifties, a controversy which not
only caused great excitement at the time, but also produced after efi^ects that
have been felt ever since. It may therefore be worth glancing at, all the more
so as the parties to the dispute were exclusively scientific investigators, some
of whom were very distinguished, while philosophical and theological
opinions do not come into consideration at all.
Among those who became involved in this dispute Justus Liebig (1803-
73) ranks first; on thewhole, he may be considered one of the greatest scientists
MODERN BIOLOGY 449
of the century. The son of a colour-man at Darmstadt, he acquired in his
father's shop even as a child an interest in chemistry and its practical applica-
tion. At one time he endeavoured to follow his bent as an apprentice to an
apothecary, but did not get on well there, and he then studied for a couple of
years at German universities, during which time he associated himself with
Schellingianism; he soon wearied of this also and went to Paris, where he
eventually found the training he sought for in the laboratory of the famous
Gay-Lussac. On the recommendation of Humboldt he was called to the chair
of chemistry at Giessen, and after struggling for years against jealousy and
hostility he succeeded in bringing into being the first chemical university-
laboratory in Germany. As a teacher he resembled J. Miiller in his capacity
for gathering around him and educating numbers of pupils; indeed, the re-
vival of chemistry in Germany is attributable to him. Towards the close of
his life he became a professor at Munich. He was a pioneer in his purely
chemical discoveries, especially in the field of organic chemistry; he gave
to organic elemental analysis the form that it has retained ever since, while
his investigations into organic acids were of epoch-making importance, as
were also his discoveries in the sphere of zymurgy. These latter discoveries
made him the foremost supporter of the chemical fermentation theory, and
Pasteur's stubborn opponent. He is of greatest importance, however, as the
creator of practical agricultural chemistry; hitherto it had been thought
generally that plants absorbed their principal nourishment out of the sur-
face-soil, but he proved that the surface-mould was rather augmented by
cultivation, that carbonic acid was the plants' sole source of carbon, and
ammonia its source of nitrogen, and to prove his theory he instituted experi-
ments with manure on an expensive scale. As a result of these experiments he
placed agricultural economy on a natural-scientific basis, but he certainly
shot far beyond the mark — partly owing to his ignorance of vegetable
anatomy — and he gained many enemies on account of his overbearing
polemic, especially against the plant-physiologists. These in their turn ex-
posed a number of Leibig's inaccuracies; he denied the value of nitrogenous
manures, he wanted to supply the earth with insoluble instead of soluble
phosphoric acid and potassic salts, and he entirely ignored the respiration of
plants. On many points he received sharp criticism at the hands of Schleiden
and Mohl. As an animal-physiologist Liebig also acquired fame for his pio-
neer studies of the preparation and utilization of foodstuffs; he ascertained the
chemical compounds that are conveyed to the body through the food, but
here, too, he often went wrong, as when he divided food-substances into
"plastic" and "respiratory," including albuminous substances among the
former, and fats and carbohydrates among the latter.
In this sphere Liebig was opposed by a young Dutch physiologist,
Jacob Moleschott (18x2.-93). The son of a physician, he studied physiology
450 THE HISTORY OF BIOLOGY
— and at the same time Hegel's philosophy — at Heidelberg and became a
lecturer there, but was dismissed on account of his "materialistic" views.
He then became professor at Turin and afterwards at Rome. He introduced
research in experimental physiology into Italy and carried out valuable
investigations, especially in the sphere of the phenomena of respiration.
These brought him into conflict with Liebig, whose theory of the influence
of food-substances upon breathing he rejected. But at the same time he made
a violent attack upon Liebig's entire conception of the cosmos. In a series of
popular papers, Chemische Briefe, the latter gave an account of the progress
of chemistry, in the course of which, confirmed Schellingian that he was, he
extolled in fervent eulogy the wisdom and might of the Creator. In opposition
to these letters Moleschott wrote a book, Kreislauf des Lebens, in which he
attacked Liebig in vigorous though courteous terms, and in connexion there-
with produced a purely materialistic conception of the world. This he bases
on the theory of the permanence of energy and on the syntheses of Wohler;
on the other hand, unlike Comte, he propounds no original ideas on evolu-
tion. To him life is a magnificent process of metabolism; thought is a product
of the activities of the brain. As a confirmed Hegelian he delights in abstract
speculations; through combining these with physiological theories he often
becomes involved in a helpless confusion of thought. Albert Lange in his
Geschkhte des Mater ialismus quotes some amusing instances of Moleschott 's
muddled attempts to get away from the contrasts between subjective mental
impressions and objective reality, and of his still more confused ideas of
matter and energy; after quoting a more than usually vague page of Moles-
chott's book, he asks: "What part of the philosophical backwoods are we in
now?" In fact, Moleschott has no idea of the limits of scientific research; in
accordance with the idealistic philosophy that he once embraced he imagines
that he can explain the whole of existence by a few artificial ideas. On the
other hand, Liebig certainly had no thought of letting natural science hold
its own and leave it to religion to satisfy the ideal requirements of life —
showing that he too was a victim of the vagueness of thought that romantic
philosophy left in men's minds.
Another important naturalist who was involved in a similar controversy
was Rudolph Wagner (1805-64). He had studied medicine and taken his
degree at Wiirzburg and afterwards worked under Cuvier in Paris, eventually
being appointed Blumenbach's successor at Gottingen. He was a creditable
investigator and teacher; among his pupils were such men as Leuckart and
the philosopher Lotze, and among his works his investigations into sper-
mato- and ovogenesis and into the tactive corpuscles are especially worthy
of mention; he was also reputed as an anthropologist, in the spirit of Blumen-
bach. At a scientific meeting at Gottingen in 1854 he gave a lecture on
Menschenschopfung und Seelensubsfan^, in which he discussed the question of
MODERN BIOLOGY 451
the origin of man from one single pair in accordance with the Church's
doctrines of creation — a question which he certainly believed anthropology
to be incapable of proving or disproving, but which gave him an opportunity
of making a violent attack upon the materialistic soul-theories of the time,
which he inveighed against from the point of view of both science and moral-
ity. He himself worked out a theory of the soul as a kind of ethereal sub-
stance, which leaves the body at death and imparts itself to the children that
are born — an idea somewhat reminiscent of Swedenborg's spirit theory. His
antagonist on this subject was KarlVogt (1817-95), who ^^'^ been professor
at Giessen between the years 1847-9, but had been removed on account of his
having participated in the revolutionary movements of that period; he after-
wards became professor at Geneva, gaining a reputation especially as an
author of sound text-books and popular scientific works. Between him and
Wagner there ensued a controversy on the question of the creation and the
soul of man, which rapidly degenerated on both sides into sheer lampoonery,
involving personal insults of the basest kind. In this Vogt maintained that
the different human races cannot have a common origin and in support of his
argument adduced a number of proofs of the constancy of species and varie-
ties, which were not quite in the spirit of the theory of the origin of species.
Further, there was considerable discussion as to the fertility of hybrids, which
Vogt upheld and Wagner denied, and finally Vogt found an easy butt for his
witticism in Wagner's divisible soul-substance, and at the same time main-
tained the assertion that the soul was a product of the brain, which "pro-
duces ideas as the liver produces bile and the kidneys urine." On the whole,
Vogt seems to have been entirely unmoved by the earlier natural philosophy;
this frees him from having to solve a number of problems that his philo-
sophically trained contemporaries felt themselves bound to take up for dis-
cussion, but, on the other hand, it involved him in gross self-contradictions.
The most painful feature of this polemic, however, was its markedly political
character; on the one hand, a Christian conservative professor, holding a
good position and boasting of his friendship with statesmen and ministers,
and, on the other hand, an exiled revolutionary, embittered by the shipwreck
of his ideals and by his own misfortunes. It would almost appear as if the
whole of this scientific controversy was merely an excuse for giving two in-
dividuals from opposite political camps an opportunity of coming to grips.
In fact, the antagonism of the two ideas, materialism and idealism, retained
this character in Germany not only during the decade with which we are
dealing, but also up to a far later period; the points of view as to the soul's
"to be or not to be" coincide with the attitude: supporter of the Government
or supporter of the opposition. During the eighteen-fifties, as we have seen,
the representatives of radical ideas at the universities found themselves in
quite a difficult position as far as regards educational freedom; this state of
45 2. THE HISTORY OF BIOLOGY
things was certainly improved at a later date, but for some considerable time
to come Christian conservatism was an officially approved standpoint.
A radical thinker who never succeeded in acquiring any permanent
right to give instruction was Biichner, one of the most widely read of the
authors who wrote on the materialism controversy. Friedrich Karl Chris-
tian LuDwiG BiJcHNER (18x4-99) belonged to a very gifted family, especially
in regard to literature. He studied medicine and concurrently also philosophy,
was a lecturer for a time, but having been dismissed, he earned his living by
taking up a medical practice. He was of noble character and a keen upholder
of liberty and justice, and from his early youth he enthusiastically adopted
materialistic ideas, in which he saw a means of bringing humanity out of
darkness and superstition. His famous work Kraft und Staff, one of the most
widely read popular scientific works of his age, is really a collection of
talks on various theoretical questions in connexion with natural science,
written in an attractive form, but without any very great originality. The
old theme — the indestructibility of energy, the permanency of matter, the
soul as a combination of cerebral functions — is played upon with constant
variations and in a spirit of incessant controversy against theologians and
philosophers. Biichner certainly has a better idea of the limitations of
natural science than Vogt; he admits that existence is full of riddles that
cannot be solved; but like Moleschott and Vogt he never attained to that
clearly formulated self-limitation that Comte in his great work imposed
upon positivism. Nor did any of them realize the importance of evolution as
Comte did. All of them hailed the advent of Darwin with enthusiasm; his
doctrine gave to their conception of nature an impetus that it never had be-
fore. The fact is, the idea of the origin of species gave to the realistic natural
philosophy the connexion that the idealistic conception of nature had in its
theory of ideas. Energy and matter were far too abstract and difficult ideas
to support a popular theory of life, all the more so as the above-mentioned
champions of their omnipotence lacked that thought-training which would
have made them capable of mastering a subject so hard to elucidate. Their
service to natural science and their labours for its propagation among a
larger public are at any rate deserving of recognition.
FROM DARWIN TO OUR OWN DAY
CHAPTER X
THE PRECONDITIONS OF DARWINISM
I. Modern Geology
DURING THE ZENITH of the power of Darwinism it was considered in
certain quarters that one of the chief missions of cultural history-
was to seek after "pre-Darwinists." It was obvious that in such
circumstances aspirants to this honour should come forward in large num-
bers; to begin with, the old Greek natural philosophers Anaximandros and
Empedocles were named, and the number increased the nearer one came to
modern times. There came another period when the list of personalities thus
accumulated could be used to depreciate Darwin, as Kohlbrugge used it.' If,
however, we damp our enthusiasm somewhat and have regard to actual
facts, we shall find that the precursors of Darwin were far fewer. He him-
self has acknowledged the influence that he derived from Lyell's geological
theories and Malthus's studies of population, and it seems only fair when
reviewing a scientist's development to take into consideration his own re-
marks on the subject. If we do this, we get two preconditions for the origin of
the Darwinian theory — a natural scientific, or, more exactly, a geological, and
a socio-political. We shall now proceed to consider the former of these two.
Compared with biology, modern geology is a young science. Some of
its pioneers have been mentioned in the foregoing: da Vinci, Steno, BufFon.
The creator of geological study as a special branch of science is without
doubt Abraham Gottlob Werner (1750-1817), professor at the mining
academy at Freiberg, a teacher of Humboldt and many other geologists and
mineralogists. He systematically explored the geology of his own district,
determined the sequence of the rock-beds, examined their composition, and
on the results thereof based a rational mining-industry. He never actually
printed his theories; it is only through the medium of his pupils that the
world has become acquainted with them. He is best known as the advocate
^ Kohlbrugge, "War Darwin tin originates Genid" Biologisches Zcntralblatt, Vol. XXXV, p. 93.
453
454 THE HISTORY OF BIOLOGY
of "Neptunism"; he believed that all mineral species, even basalt, are pre-
cipitated in water. The narrowness of his conclusions was largely due to
the fact that he never made any journeys; he presumed that the geological
conditions all over the world were like those in his own country. The energy
with which he defended his views was, however, impressive, and his pupils,
who came from all parts of the world, endeavoured faithfully to apply the
master's doctrines, however difficult they might prove to be in practice. The
whole of the earliest generation of geologists, as a matter of fact, shared
this failing of Werner's — even the scientist who is named with Werner as
the creator of geology, Hutton, had never been outside his own country.
James Hutton (172.6-97) was the son of a Scottish landowner, and
studied medicine in his youth, but, having inherited a fortune, he after-
wards devoted himself entirely to scientific research, especially geology. It
was not until late in life that he published the work Theory of the Earth, in
which he expounds his original ideas, though in a not very clear form. He
considers that geology has nothing to do with the history of creation; its
function is to describe the rock and earth strata now existing and to account
for their origin. He believes that the present rock-beds have arisen through
the destruction of older strata, similar to that which takes place daily
through the influence of water. This principle of explaining the past out
of the present represents his most valuable contribution to the development
of geology, though his own applications of that principle were often not
very successful.
It was not possible to ascertain the reciprocal age of the different rock
strata, and thereby also to create a history of the evolution of the earth's
surface, until attention had been paid to the remains of living creatures that
are found in the various geological beds. This, indeed, Buffon had already
done, but the one who really systematized palaeontology was Cuvier. His
work in this sphere has already been described and its deep significance
pointed out; his catastrophe theory, the gist of which has likewise been
explained above, had disastrous consequences. Its influence was felt least in
England, where geology was developed independently in this field also. The
scientist who introduced into that country the knowledge of fossils as a
guide to geological research was William Smith (1769-183 9). Born of poor
parents in the country, he received a deficient school-education and after-
wards became apprentice to a surveyor, who taught him sound professional
knowledge, with the result that he was sought after as a surveyor and level-
ler, making a fortune in that profession and at the same time having oppor-
tunities for studying very different geological strata and rock formations.
He quickly came to realize that these possessed a settled order of succession
and that different animals and vegetable remains characterize the different
stratifications. The fossils he himself was unable to determine, this being
MODERN BIOLOGY 455
done by some of his friends, but he had a keen eye to the place into which
each form should fall in the strata system. Eventually he published the
results of his life-work in a great geological atlas of England, which cost
him his whole fortune. For a time he suffered want, but was eventually
granted a government pension, which ensured him a peaceful old age. It
was through him that the use of guiding fossils for identifying the age of
geological formations was introduced into science.
An investigator who surveyed, and in a high degree developed, the
geological knowledge of his age was Christian Leopold von Buch (1774-
1853). He belonged to a distinguished and wealthy Prussian family, and
studied under Werner at Freiberg together with Humboldt with a view to
entering the mining service, but he soon applied himself entirely to geology,
which, thanks to his inherited wealth, he was able to study without having
to earn his living. He made extensive expeditions, in the course of which
he made a particularly fine collection of comparative material from various
countries. The result of this research work soon led him from the Neptun-
ism of Werner to the opposite extreme; he ascribed to volcanic activity an
important, and indeed far too important, part in the history of the earth's
surface. His investigations, carried out in different regions, are nevertheless
of lasting value; he w^as, moreover, an eminent palaeontologist, making
valuable investigations of special subjects, particularly of fossil inverte-
brates: Cephalopoda, Brachiopoda, and others.
Charles Lyell is, however, the scientist that is first worthy of mention
as the founder of modern geology and thereby as a pioneer of the descent
theory. He was born in 1797, the son of a Scottish landowner, who was also
interested in botany and who inspired in his son a passion for nature study.
The latter took his degree at Oxford and afterwards adopted the profession
of a lawyer. But he did not go far in that career, for eye-trouble compelled
him to give up public work of any kind. Long before this, however, geology
had attracted him, W. Smith's investigations especially interesting him, and
for the rest of his life he devoted himself to that study, bearing with un-
paralleled courage the severe deprivation that defective vision always means
to a scientist, especially a natural scientist. One source of comfort in these
circumstances was the fact that his wife with devoted self-sacrifice dedicated
her life to helping him in his work. He thus became one of the many bril-
liant private scholars in which the cultural history of England abounds, and
he was the recipient of not a few honours. He undertook a number of long
voyages of exploration; he considered them to be indispensable for a geol-
ogist, for it is only thus that he can gain that living idea of the various
forms of the earth's surface which may serve as a basis for a theory of its
history. The rest of his time he spent in London, where he was a member of
many learned societies and was also otherwise held in high repute. He died
45 6 THE HISTORY OF BIOLOGY
in 1875, ^^^^^ having some years previously lost both his sight and his wife,
who had been the mainstay of his life and his work.
Ly ell's act uali Stic geology
Like Hutton, Lyell takes as his starting-point the present form of the earth's
surface, studies its changes as the result of various natural influences, and
finally draws the conclusion that the same forces have always, and approxi-
mately in the same degree as in our own time, been operating on the earth's
surface; he who declares otherwise must substantiate his argument with
proofs; it is the upholders of the catastrophe theory whose duty it is to
prove the correctness of their views, and not vice versa. This conception of
the evolution of the earth — it has been named the "actualistic" — forces
Lyell to follow it through to its extreme consequences and far beyond what
science in modern time is prepared to admit. Thus, in his Principles of Geol-
ogy he absolutely denies the possibility of the earth's having originally
existed in an incandescent state; he likewise definitely rejects Lamarck's
theory that the animal world in earlier ages consisted of entirely different
species from those in modern times and declares that mammals and birds
have existed from the very earliest times. But apart from these extravagant
statements, which as a matter of fact he afterwards partly corrected, his
strict adherence to the principle that the phenomena of past ages should
be explained from what is known from the phenomena of the present time
has formed the basis on which it has been possible to construct a truly
scientific geology. The earlier geological theories, both ingenious and fool-
ish, had all been mere products of the imagination; Lyell introduced the
principle, which must inevitably be adopted by every empirical science, of
starting from what is known and has been investigated and thence pro-
ceeding gradually towards the more remote and the unknown. If past natu-
ral phenomena in general are to be calculated or at least reconstructed with
fair probability, it is necessary to start from the present, whose course
of events it is possible to survey. This astronony has long done with its
calculations of the position and motions of the heavenly bodies in past ages;
and modern geology has in certain spheres, as, for instance, in the deter-
mination of annual stratifications out of water during preceding periods,
reached a degree of accuracy that should not be far inferior to astronomi-
cal calculation. And this principle essentially represents Lyell's service to
science.
His criticism of La7narck
Moreover, Lyell has made important contributions in his above-mentioned
work to problems of the development of life upon the earth. His criticism
of Lamarck's theory undeniably touches the latter's weakest spot, when he
maintains that Lamarck never even attempted to find out the origin of a
single vital organ, but merely occupies himself with modifications in those
MODERN BIOLOGY 457
already existing. He can hardly be blamed for the fact that he does not con-
sider that he had found any actual proof of the transition of one species to
another, since it has indeed scarcely been possible to discover one even in
our own time. He does not believe in the possibility of the various species'
being able to vary beyond a certain limited extent, and this limit is soon
reached; if we try to force a form beyond this, it perishes; as an instance he
quotes the adaptability of species to different climates. Man's domestic ani-
mals have from the beginning been especially suitable for taming, while
other equally or more intelligent animals, the apes, for instance, have to
be left at liberty. It is primarily, however, the rare existence of and sterility
in hybrids that to Lyell's mind gives proof of the constancy of the species.
The similarity between embryos of various kinds merely testifies to a com-
mon plan in their structure, but no common origin. He believes that every
species has been created in a locality suitable to it and has spread from there
under the constant influence of the climate, means of subsistence, and com-
petition with other life-forms. In disproof of Lamarck's theory of species-
modification he maintains that an alteration in the climatic changes or other
alterations in the conditions of life would give certain species advantage
over others, so that the adaptability assumed by Lamarck would never be
realized in the latter. If a lake were to be converted into a swamp, already
existing marsh-plants would be ready to overrun its area, while the aquatic
plants would die out before they had time to adapt themselves to swamp
conditions. How the species came to be created is a question that Lyell
refuses to discuss; he speaks of "creative force," though he attributes no
personality to it, regarding the whole problem as insoluble. Instead he dis-
cusses in detail the conditions governing the distribution of species, their
development and extinction during different geological epochs. The whole
of this exposition exercised a very great influence on Darwin, both positive
and negative, by calling forth a contradiction from him — a point on which
more light will be thrown later.
But the main point is that Lyell's theory of geological evolution offered
at the time particularly valuable support to the idea of evolution, which
was one of the watchwords of the age; here indeed there was confirmation
in nature herself of the idea of an uninterrupted development as the funda-
mental force in existence. The result was that Lyell's name became one of
the most popular at the time, and he himself enhanced his reputation by
his ability to keep pace with scientific developments; he, the opponent of
Lamarck, associated himself directly and without reservation with Darwin.
His activities as the promoter of Darwinism will be dealt with in the next
section.
458 THE HISTORY OF BIOLOGY
X. The Ideal Preconditions of Darwinism
Failure of Lamarck' s theory
The question has not infrequently been discussed: Why did not Lamarck's
theory of evolution succeed? The reason has been put down to the opposi-
tion of the Church, but certainly without justification; as far as is known,
Lamarck was never interfered with by the Church, and the latter's opposi-
tion to the theories of origin and species-m.odification is, as we shall find,
of a far later date. We are, then, far more justified in blaming the romantic
natural philosophy, which, seeking, as it did, after one common idea for
every life -form, lacked all feeling for material development. For herein lies
the real gist of the problem: if a theory of evolution is to attract general
attention, there must naturally be evinced an interest in evolution. Our next
duty, therefore, will be to try to explain how this interest arose and how it
expressed itself at the time of Darwin's appearance.
It is common knowledge that mankind is always ready to fix its ideals
in antiquity — "the good old times." One is most inclined to deplore the
present and to view the future with feelings of anxiety. And just like indi-
viduals, the public opinion of the different epochs has done the same; if
man has carried out reforms, it has mostly been done under the form of
reviving the ideal conditions of ages long past; so it was during the Refor-
mation, when people vv^ished to revert to the conditions of early Christianity,
and so too during the French Revolution, when people raved over the re-
publics of antiquity, and imaginative popular leaders called themselves An-
acharsis or Gracchus. If one has dared to cast a glance at the future, one has
most probably expected to find happiness in some vast catastrophe resulting
in the total annihilation of the present; thus all apocalyptical enthusiasts
of antiquity and ever since, and thus too the political extreme tendencies of
modern times. Belief in a gradually progressive, law-bound development has
always been limited to a few, and these perhaps are to be found among the
men of action rather than men of thoughts and words. The most pronounced
faith in progress that has ever existed has been the liberalism of the nine-
teenth century, a current of ideas which had just reached its zenith by the
middle of the century, when the theory of origin came to the fore. The co-
incidence is of course not accidental; on the contrary, the one idea is de-
pendent on the other, and therefore the victory of Darwinism is inexplicable
without some insight into the general intellectual conditions at the time
of its birth.
Liberalism of the nineteenth century
The optimistic belief of liberalism in the progress of the human race had its
true origin in England, where throughout the entire eighteenth century
prosperity and enlightenment increased slowly but surely, where humane
Pi
<
Q
<
u
^^
Si
i
^•'^^
^ ''^^BSBBL I ^
iW .^1
B^^
v^fii^^^H
■:^BQH^^^^H^^^^^^BB j
|jj|i w
^■hq
1
^^<<^^,< ■v\^IHp
W^.:': * . I .^
lir ?
iSte.^
r^ /(
"^■•^=1
IL
^^B-
^V^HHBI^>'
I
PC]
1/3
w
<5
u
MODERN BIOLOGY 459
legislation and democratic social development were demanded and even grad-
ually achieved without violent upheavals. This beliefwas strongly influenced
by Rousseau's doctrines of the natural goodness of man; which has only been
perverted by social life and by the oppression of evil kings and priests; it
found expression in the democratic reforms of the French Revolution, but
it acquired its true character through the great technical and material prog-
ress made during the nineteenth century, to which reference has already been
made above. The new big-scale industrial development and world trade, ren-
dered possible by steam-power, created an intelligent middle class, which
felt well satisfied with the present and hoped for still greater benefits from
the future; the labouring classes were not yet organized and their discontent
was thus perceptible only in isolated instances. The vast production of ma-
terial values set its mark upon the age and was met by the belief, adopted
from Rousseau, in the natural goodness of the human race and in Bentham's
doctrine of happiness for as many as possible as the chief aim in life; hap-
piness was made the synonym for material welfare, and this could best be
attained by letting mankind, endowed by nature with goodness and intel-
ligence, look after themselves, undisturbed by oppression and superfluous
regulations. Human life thus came to be regarded as a dominion of imper-
sonal forces guiding humanity with the necessity of a natural law towards
better times, if only they were allowed to operate freely. The people — the
impersonal summary of the individuals living in a country — were better
advised than any single person; if only they were allowed to look after
themselves, their activities would conduce to a successful development, to
which there seemed to be no limits. Free competition both in the material
and in the spiritual world and no interference with the individual's liberty
of action were the watchwords of the age; how the free will of the indi-
vidual was eventually to be reconciled with the popular will was a question
that did not bother the minds of many; for the time being, the individuals
looked up to the popular will as to a higher power, the only fault of which
was that it had not yet had sufficient time in which to operate.
This conception of life, which naturally appeared under quite different
forms in different quarters — in historians like Buckle, in thinkers like Mill
and Spencer, not to speak of their pupils and imitators on the Continent —
was without doubt the most favourable soil possible in which to cultivate
a general theory of evolution. Evolution, Progress, were in fact the slogan
of the age. It had been employed in Comte's system, described above, but
only as far as regards human culture; through Darwin evolution was ele-
vated to a natural law governing all life. It is no wonder, then, that his
theory was hailed with enthusiasm by all those who cherished the ideals
of the new age. It was indeed the ideal itself that was hereby sanctioned to
embrace the whole of nature; on the other hand, it affords an explanation
460 THE HISTORY OF BIOLOGY
for the violent opposition on the part of all adherents to the old order of
society, which was not yet won over and which towards the close of the
century was to muster increasingly stronger forces. But Darwin himself was
influenced by the new conception of community life; it was from one of its
theorists that he obtained the actual idea for his theory of selection —
namely, Malthus, wherefore he is worthy of a place in the history of biology.
Thomas Robert Malthus (1766-1834) was the son of a landowner,
took his degree at Cambridge, was ordained priest, and obtained a curacy,
but at the same time devoted himself to the study of national economics.
On account of his writings he was given a professorship in London, where
he afterwards worked with great success. His father had been a personal
pupil of Rousseau and entertained rather radical views on the improvement
of the human race by a fair distribution of wealth — an idea which had then,
as it had later, many supporters. Against them there appeared the younger
Malthus with his chief work. The Principle of Population, which came out
in many editions and has been very widely discussed. Although himself a
liberal, he is in no wise a revolutionary optimist; he sees the cause of human
misery not in an unfair distribution of property, but in man's own habit of
living thoughtlessly and frivolously. The cure for this he sees in bringing
mankind up to exercise self-control, every man being taught not to raise
a family without definitely guaranteed means of subsistence. It is, he be-
lieves, a fact throughout nature, in plants, animals, and human beings, that
natural procreation is stronger than the possibilities of maintaining life;
from this there arises in nature a violent competition for the maintenance
of life, and in human life there is, further, helpless and ever-increasing misery
among the poverty-stricken classes, which no philosophical measures can
remedy. He then tries by means of historical and geographical-statistical
investigations to find out how it is that the increase in the population never
has followed, and does not follow now, its natural course, but is restricted,
more or less owing to the fact that the supply of maintenance is limited,
with the result that want, with its concomitant vice and crime, thins out
a great number of the poorest in each community. Upon its first appearance
this doctrine was violently opposed from both conservative and radical quar-
ters; it is not, however, within the scope of this work to enter into a detailed
discussion of the subject; it is sufficient to point out the above-mentioned
theory of competition, which gave Darwin the idea for his theory of selec-
tion. To this latter we shall now proceed.
CHAPTER XI
DARWIN
CHARLES Robert Darwin was born in 1809 at Shrewsbury in the west
of England. His father, Robert Waring Darwin, was the son of the
physician and natural philosopher Erasmus Darwin and was him-
self a physician. He was married to Susannah Wedgwood, daughter of the
famous procelain manufacturer Josiah Wedgwood, who, from being a poor
and ignorant apprentice to a potter had made a successful career, acquiring
a splendid fortune and a famous name in the history of ceramics. Charles
was the sixth out of eight children. He went through a school of the usual
English type, his education consisting almost exclusively of the classical
languages, and was afterwards sent to Edinburgh in order to study medicine
in the family tradition. The Latin he learnt at school did not interest him
very much and he was utterly bored by the anatomy lectures. Darwin broke
off his medical studies after a couple of years, so that he never became an
anatomist, to his own great loss. He now decided to try his hand at theology
at Cambridge, where he spent three years and took his degree of bachelor of
arts, but he spent most of his time pursuing the usual occupation of the
well-to-do English undergraduate — sport, especially shooting. He also col-
lected insects and plants for his own amusement, but he chiefly interested
himself in geology, receiving a sound elementary training in that subject
under the guidance of the eminent professor Adam Sedgwick (1785-1873),
whom he accompanied on several expeditions. On the recommendation of
a friend he was offered in 1831 the unsalaried post of naturalist on board
the cruiser Beagle, which was to circumnavigate the world for mainly carto-
graphical purposes. This voyage, which lasted five years, gave him, as he
himself says, his real training as a naturalist, as it also determined the di-
rection that his future work was to take. He worked with zeal and sent home
from the various stopping-places on the way both notes and collections.
Of these the geological possessed the greatest value; the zoological and
botanical were regarded by contemporary judges as nothing extraordinary.
This persevering activity was so much the more praiseworthy as Darwin
suffered throughout the journey from incurable seasickness, which gradually
irremediably impaired his health. On his return home he devoted himself
for years to the working up of the natural objects and the material for ideas
that he had gathered in the course of the voyage. During that period there
461
46z THE HISTORY OF BIOLOGY
slowly developed in his mind the theory which bears his name. In 1833 he
married his cousin Hannah Wedgwood. Her wealth added to his own made
it possible for him during his remaining years to lead the quiet life of a pri-
vate scholar, which in fact became in time an absolute necessity, owing to
his increasing ill health. Three years after his marriage he left London and
settled in Down, a small town in Kent, where he spent the rest of his life
in his own comfortable house, with a delightful garden. Even in these cir-
cumstances, however, his health did not improve; he suffered from a nerv-
ous stomachic trouble, which occasioned constant vomitings and frequent
insomnia. It was only through living a painfully regular life under the self-
sacrificing care of his wife that he was able to hold out as long as he did.
His days passed with brief but intensively concentrated periods of work, al-
ternating with medical attention, walks, and literary diversion; journeys
and social life were restricted to a minimum. During this period there was
given to the world that unique production — considerable even in its extent
— which made his name immortal. His bodily existence, so full of suffering,
was compensated for throughout his life by a rare spiritual poise; complete
freedom from passion, from hate, envy, and ambition, and an almost tender
amiability, which certainly found it difficult to refuse a petition, however
unreasonable, but which also made it easy for him to enjoy and find child-
like pleasure in the narrow life to which his ill health restricted him. His
was no critical character; towards the statements of others he used to show,
as Johannsen says, "an amiable credulity," and his own experiments were
often consciously childish. His sensitiveness, however, was in no way as-
sociated with weakness of character; on the contrary, few students of nature
have striven with such unbending determination for years and years towards
a given goal, and adhered to a point of view when once adopted with such
firm conviction. His ideas were, as is well known, both unreservedly praised
and violently vituperated; attacks were met by him with unfailing stead-
fastness and a noble calm, so that he never allowed himself to be involved
in personal polemics, but he always took note of and parried material ob-
jections. Thanks to these qualities, Darwin came in the course of years to
enjoy personal esteem such as seldom falls to the lot of scientists. Occupied
in constant work, his life moved quietly towards its close. He died in i88z
and was buried in Westminster Abbey, not far from Newton, followed to
the grave by the most distinguished men in the country both in the social
and in the scientific world. Shortly before his death he had wTitten down in
some notes on his own life the oft-quoted words: "As for myself, I believe
that I have acted rightly in steadily following and devoting my life to sci-
ence. I feel no remorse from having committed any great sin, but have often
and often regretted that I have not done more direct good to my fellow
creatures."
MODERN BIOLOGY 463
In his youth Danvin was a confirmed lover of the open-air life; a good
shot, an enthusiastic huntsman, and a keen observer of life in nature. This
love of animate life in the open air he retained even in his old age; long
after ill health had compelled him to give up shooting and voyages of ex-
ploration, he applied himself with indefatigable devotion to the care and
observation of life in his park and garden. Dogs and cats, birds, insects, and
earthworms, no less than plants of the most varied kinds, were to him a
never-wearying source of joy and observation; all their manifestations of
life in the minutest detail were the object of his most careful study; animals'
actions, instincts, and manifestations of intelligence were observed, analysed,
and summarized by him day by day and year by year with never-failing in-
terest. His theoretical training, on the other hand, was deficient — -most thor-
ough in the sphere of geology, whereas in biology it was, on the whole,
limited to the systematic side. His observations made during the circum-
navigation of the world also bear witness to this restricted basis on which
his education was founded. He was, moreover, in his youth a firm believer
in the Christian faith — he intended, in fact, to become a clergyman — and he
accepted without criticism the traditional dogmas, including, of course, the
doctrine of the origin of living species as the result of a divine act of crea-
tion. During his voyage, however, he found that this belief conflicted with
the results of his observations. His diary contains many proofs of this; in
particular, the existence of many species with a small area of distribution,
of forms closely allied to one another, but not alike, and taking the place
of one another in different localities, yet not existing together, seemed to
him difficult to reconcile with "nature's great plan." Why had it been neces-
sary to create all these slightly differentiated and narrowly distributed spe-
cies? He spent one month on the desolate Galapagos Islands, situated a long
way off the coast of South America and composed of volcanic lava compara-
tively recently cast up out of the ocean; here he felt himself "placed in prox-
imity to the very act of creation itself." But here he found a fauna of markedly
South American genera, though possessing peculiar species; of many birds
each separate island had its own species. That one species should have been
created for each small island seemed to him irrational; but how, then, had
the different species arisen and why did they belong to the South American
genera? This problem, having once penetrated his mind, gave him no rest.
Upon his return home he at once started to record in a separate book his
experiences in connexion with the question of the formation of species, and
he sought long and restlessly for proofs of the correctness of his ideas. In
1844 he writes in a letter to his friend the botanist Hooker: "I have read
heaps of agricultural and horticultural books and have never ceased col-
lecting facts. At last gleams of light have come, and I am almost convinced
(quite contrary to the opinion I started with) that species are not (it is like
464 THE HISTORY OF BIOLOGY
confessing a murder) immutable." Lamarck's theory of the modification of
species, however, Darwin was unable to accept; it appeared to him to be
"rubbish" — "Heaven forfend me from Lamarck's nonsense of 'a tendency
to progression. ' " Nor indeed in any other biological literature accessible to
him could he find any way out of the difficulty involved in the origin of
species.
Darwin s experiments to -prove the mutability of species
During this period he was closely associated with Lyell, the scientist who
most influenced him — he too, as we have seen, no friend of Lamarck —
and resolved to deal with the species as Lyell had dealt with the geological
strata of the earth — namely, to collect as many facts as possible regarding
the transition from one form to another. In this respect domestic animals
seemed to him to give the best suggestions : that each separate domestic animal
was a true species no systematist had ever denied and it was likewise ac-
knowledged that man had produced a mass of different forms of every species
of that kind. Darwin placed himself in communication with a great many
animal-breeders, and himself for years bred different races of pigeons, all
for the purpose of discovering how the different races arose. Expert breeders
believed that by a selection of suitable parents it was possible gradually to
modify the progeny at will. Darwin also came to accept this view; all the
young in a litter of domestic animals are indeed somewhat unlike one another
and their parents — they "vary" as he says — and by selecting the suitable
variations it is possible to guide the breed in the required direction. But if
man was able by selection to produce out of the uniform canine type that
still exists among wild tribes such a large quantity of different forms, should
it not then be possible for species to be modified by nature in the same way?
The difference between a greyhound and a bulldog is far greater than that
between many wild life-forms which without doubt pass for good species.
But is there in nature a force operating in the same direction as the breeder
when he selects new forms of domestic animals? Here lay the worst stum-
bling-block. Then Darwin happened to read Malthus's above-mentioned
work on population : how both in nature and in human life there are produced
individuals in far greater numbers than there are means for maintaining, and
how the weakest perish in the competition for food. This gave him his idea;
in the struggle for existence those life-forms are destroyed that are least ca-
pable of adapting themselves to prevailing conditions, while the strongest
individuals survive and reproduce those qualities that have a greater chance
of survival. Thus the external conditions themselves come to multiply the
differences brought about by the variability of the offspring in relation to
the parents, until new varieties and new species arise. Consequently, the
struggle for existence induces a natural selection that operates similarly to
the choice of races among domestic animals exercised by man, only with
MODERN BIOLOGY 465
this difference — that vast expanses of time are available for natural selec-
tion, which justifies the assumption that all the manifold forms of life on
the earth, both those which have existed and those which still exist, have
been developed through its influence. These facts, then — the dissimilarity
between the offspring and the parents (that is, variability) and the struggle
for existence, with the resultant natural selection — explain, according to
Darwin, the origin of species.
His Zoological works
For two decades Darwin kept this theory to himself in an unceasing search
for fresh proofs of its universal application. Finally, in 1859, he published
it in a work entitled On the Origin of Species by Means of Natural Selection, one
of the most famous works of natural history that have ever been written.
Even before that, however, he had won a high reputation as the result of
a number of monographs on various subjects. Among these may be mentioned
two geological treatises: On Volcanic Islands and On Coral Reefs, the latter
being specially famous for its universally accepted theory of the arising of
atolls or circular reefs through the sinking of the land area around which
the coral reefs had originally grown up. Among his zoological works may
be especially mentioned an extensive work on the Cirripedia, in which he
gives a detailed and exhaustive description of the system and evolutional
history of these animal forms — their peculiar dwarf males discovered by
him — besides which he has also dealt with the fossil forms of that animal
group. Moreover, as editor of the scientific results of the Beagle expedition
he contributed much of great value. It was thus a naturalist with a good
reputation who came forward with the work on the origin of species. The
violent controversy that it occasioned brought immediate world-wide fame
to its author. A somewhat detailed account of the main ideas of the work is
therefore called for, all the more so as, in spite of its immense popularity,
it would seem to have been less widely read in recent times than one might
suppose, and the exposition of the theory of origin to be found in the usual
text-books has been strongly influenced by that comparative morphology
with which Darwin himself was more or less unfamiliar.
The theory of origin that Darwin created is decidedly characterized by
the personality of its founder. Darwin brought to his work, as we have
observed above, a deficient theoretical training, particularly in the sphere
of anatomy, an intense geographical and systematical interest, and, as a
standpoint beyond which he had already advanced some way, a somewhat
ingenuous orthodox-Christian belief in the creation. Being a systematist, he
saw in the problem of species the central point of biology, and to him the
centre of this problem was, in its turn, the problem of creation. This must
be borne in mind if we are to understand Darwin's relation to the earlier
morphologically inclined generation of scientists of the Cuvier school.
466 THE HISTORY OF BIOLOGY
Whatever their view of life, the species idea was to them essentially a prac-
tical basis for comparative morphology, whereas the problem of creation
was a question that was entirely put aside as not concerning natural science. It
must at once be admitted that it was certainly due to Darwin's dilettante
conception of nature that he thus adopted just the problem of creation as
a starting-point for many years of research work and cogitation; on the
other hand, it was his treatment of the problem that caused such a public
sensation over his work.
Immutability and creation
LiNN^us in his youth defined the species as the progeny of those animals
that had been created in the beginning; he afterwards altered his view, in
that he assumed a few species to have been created, out of which the others
were evolved at a later period. To the systematists who succeeded him it
was the immutability of the species that was the essential point, the actual
basis of the system, while the problem of creation was seldom discussed.
De CandoUe, it will be remembered, has a definition of species based on
mutual similarity and fertility between the individuals, but without any
mention's being made of the creation. To Darwin, however, "immutable"
and "created," in regard to species, are inseparable terms; doubt of the im-
mutability of the species is induced by doubt of the creation, which in its
turn has been caused by the species' conditions of distribution and not by
any doubts as to the assumption of a supernatural act of creation being in
itself an explanation of nature.^ Then he gets the idea of the variations
which by means of natural selection are adapted to prevailing external con-
ditions, thus giving rise to, first of all, new varieties, and then new species.
Even earlier systematists had taken it for granted that varieties are produced
by external conditions, flourish, and disappear; what is novel in Darwin's
theory is that the species are nothing but more fully developed varieties,
which selection, resulting from the struggle for existence, has determined,
while the intermediary varieties, as being less capable of defying competi-
tion, have perished. He adduces a great many arguments to prove that those
species which are most widely scattered and, where they exist, are richest
in individuals are also those which produce the most varieties, which in
his view is the same as initial species. And he points out how vague the
boundaries between species and varieties really are in the minds of different
systematists and how difficult it is to define what is meant by species. He
considers that this name is given arbitrarily and for the sake of convenience
to a number of individuals which highly resemble one another, and that it
^ In his diary of the voyage Darwin in one place explains the absence of certain fossils
in a geological deposit by assuming that animals of that kind had not been created at the time
when the deposit came into being (Life and Letters, II, p. i). Again, in the Origin of Species a
Creator is mentioned as the ultimate cause of life.
MODERN BIOLOGY 467
is not essentially different from the term "variety," which is used for less
distinct and more iluctuating forms.
Variations in progeny
The causes of these variations, which by means of selection — natural in
wild life, human in domestic animals — are developed into varieties and
species, involve a problem that occupied the mind of Darwin a great deal.
He at once points out that only hereditary variations have any significance
and also that the essential causes of them have never been really ascertained.
On this question he adopts a somewhat hesitant attitude; it is true, he as-
serts, and collects ample material to prove, that external conditions pro-
duce variations in the progeny, which is a view strongly reminiscent of
Lamarck, but, on the other hand, he definitely rejects all Lamarckian ideas.
As a matter of fact, this theory of the heredity of variations is the basis
of the Darwinian theory, but it is also one of its weakest points; in this
connexion modern research into the problem of heredity has passed severe
judgment on him — often indeed unfairly severe, it being forgotten that he
had not that accumulation of facts to build upon which is available in modern
times, but here he certainly does touch upon extremely vague conceptions,
which make the chapter on the law of variation difficult to comprehend.
Among the circumstances that influence the individual's reproductive or-
gans and thus affect the offspring, he mentions climatic conditions of vari-
ous kinds and alimental conditions, as well as the correlation between
different parts of the body. Nevertheless, he always insists upon the im-
portance of natural selection as being greater than the direct influence of
environment. For instance, a number of insect forms on islands in mid-ocean
have restricted powers of flight as compared with their relations on the
mainland; this has arisen through the fact that those specimens that are
best at flying have been blown out to sea and perished, while the wxaker
fliers have continued to propagate, rather than through the animals' not
having dared to use their wings, with the result that their growth has be-
come stunted. Correlation, again, compels other organs to follow suit when
one organ has been modified as a result of selection. Further, he holds that
parts of the body that have become especially developed in one species, as
compared with corresponding parts in closely related species, are liable to
peculiar variation; thus, the length of the arms of the orang-utan varies,
just as, in general, every strongly developed characteristic indicates strong
variation in the previous generation. On the other hand, the wings of the
bat do not vary, abnormal though they are in comparison with the extremi-
ties of other mammals, for the entire group has wings of the same kind;
the law would hold good only if one species had longer pairs of wings than
other species of the same genus. On these grounds he believes also that
species-characters vary more than genus-characters, but the variations of
468 THE HISTORY OF BIOLOGY
Species of the same genus are analogous. In connexion with this point he
accounts for what he calls the "tendency to reversion" — the frequent and
unexpected tendency, especially in domestic animals, for forms to arise hav-
ing the characters of the wild species: tame pigeons resembling the wild
rock-pigeon, horses with zebra-like streaks, and other similar instances.
Difficulties of the evolution theory
Darwin having thus sought to determine the laws of variation, he takes
up for study the difficulties offered by the theory of evolution by means of
natural selection. The chapters devoted to this task comprise more than
half the book and represent a strange miscellany. As a matter of fact, Darwin
acknowledges no limitations to his duty to answer all objections to his the-
ory, and he always finds some way out of a difficulty, however desperate it
may at first appear. He himself considers that the most difficult phenomenon
to explain according to the theory of selection is how the ants' workers
have acquired their intelligence; they cannot reproduce themselves and thus
transfer their favourable variations by heredity to any offspring. The diffi-
culty is solved by the assumption that here it is actually the community as
a whole which derives the advantage from variations; the sex-individuals
that have produced workers with the best and most advantageous qualities
have been victorious in the struggle for existence, and thus have arisen both
the highly cultivated worker types and the strongly developed instincts to
make slaves, tend aphides, etc. Darwin undertakes another particularly dif-
ferent task in seeking to explain how such a complicated organ as the eye
came to be formed. This explanation, which was ill received by contempo-
rary critics, is certainly rather far-fetched; there is no direct transition be-
tween the vertebrate animals' type of eye and that of the Arthropoda — it is
not stated why association is not sought with the molluscs instead, in which
order the highly developed visual organs of the ink-fish might have served
as a transition. — And so the whole work concludes with some general
assurances as to the metamorphosing power of selection. It is much easier,
of course, to explain the origin of the lungs from the swimming-bladder;
on this subject earlier comparative-anatomical observations have been avail-
able as a basis of study. Darwin even undertakes to defend the old objection
of the sterility of hybrids, which has so often been brought forward in favour
of the constancy of species. He differentiates between infertility as the result of
crossing species on the one hand, and the sterility of hybrids on the other;
as far as the sterility between the species is concerned, he finds that it varies
greatly in different organisms — Koelreuter's and Gartner's experiments
are especially cited as instances — ■ and the final conclusion is that "accidental
and unknown circumstances" are the cause of it in the different cases. The
sterility of hybrids, again, is compared with the infertility of wild animals
in captivity; each is attributed to the direct influence of external conditions
MODERN BIOLOGY 469
Upon the sexual organs. Here, too, reference is made to the varying results
to which different experiments have led. The fertility of variety-crosses, on
the other hand, is attributed to favourable conditions of variations in closely
related characters. The result of this is a proof that transition forms exist
between species and varieties.
Darwin and Mendel
If we compare these discussions of Darwin's on heredity and hybridization
with the experiments that Mendel concurrently carried out for the same pur-
pose, the English scientist naturally gets left hopelessly behind — on his
part, widely vacillating speculations; on thepart of Mendel, clearly conceived
and exact experiments. The very starting-point brings this out clearly;
Mendel starts from a few simple and easily determined characters and es-
tablishes their appearance in different generations in various combinations;
Darwin, on the other hand, starts from the ideas of species and variety —
that is, from the most abstract terms in biology and the most difficult to
define. In fact, in this starting-point lies the whole weakness of Darwin's
research work and speculation. His successors, indeed, almost immediately
abandoned this standpoint and instead sought for proofs of their theory by
recourse to the material and methods of comparative anatomy; Cuvier and
his successors had already studied the changes undergone by one and the
same organ in a series of diff'erent animal forms. It was through this com-
parative method's being placed at the service of the theory of origin that
Darwinism, especially through Gegenbaur and his school, came to use for
purposes of investigation objects of a definite and concrete nature. But Dar-
win himself had but little mind for comparative anatomy; he certainly cites
for the purposes of his theory a number of proofs derived from morphology,
but in quite a brief and summary fashion. He was more interested in embry-
ology. Although he himself had never worked practically as an embryologist,
he nevertheless realized the value of comparative investigations into diff'erent
stages of development and he works out the basis for a "biogenetic prin-
ciple," which Fritz Miiller and Haeckel only had to supplement.
Darwin on questions of geography
Darwin is, however, far more at home in the sphere of geology and geog-
raphy and he firmly rejects any attempts at employing the results of these
sciences to disprove his theory. The incompleteness of palasontological re-
mains he considers to be sufficient argument against those who inquire after
"missing links" between now existing genera and species, while the con-
ditions of distribution of living creatures seemed to him from the very out-
set to be the surest guarantee of the truth of his doctrine. Climatic changes
have in the course of the ages given the most powerful impulses both to new
variations and to the struggle for existence under new conditions, while
newly-formed natural barriers, mountain ranges and encroachments of the
470 THE HISTORY OF BIOLOGY
sea, have split up uniform groups of life-forms and created isolated areas of
development with accompanying new forms of genera and species. And
always variation and natural selection are sufficient to explain all phenom-
ena; since Aristotle produced his explanation of nature, no biologist has
ever conceived it his duty to the extent that Darwin did to explain anything
whatsoever. In this respect he takes the most extraordinary trouble to achieve
his purpose; in a letter to Lyell he expresses surprise that the bats of New
Zealand — of old the sole representatives of the mammals on the island —
had not made their home on the ground and developed into land-animals,
seeing that they had no competitors. And he ascribes to selection the most
remarkable powers; a traveller had seen a bear swimming in a North Ameri-
can river and snapping at insects in the water; Darwin thinks it not impos-
sible that, if food of this kind were abundant and there were no competitors,
a number of bears would become aquatic animals and would gradually
acquire larger and larger mouths, eventually becoming as monstrous as
whales. This strange conclusion, which is given in the first edition of the
work, but was modified in succeeding editions, gives striking evidence of
another weakness in Darwin's speculation: his lack of sense of a law-bound
necessity in existence. "I believe in no law of necessary development," he
expressly declares. The variations are certainly guided by laws, as mentioned
above, not, however, in any given direction, but in all possible directions,
and they are influenced, depending upon every chance, quite incalculably
by natural selection. But if, then, natural selection were guided by chance,
it would exclude the possibility of any law-bound phenomenon in existence.
Herein really lies the greatest weakness of the Darwinian doctrine of selec-
tion. It has, in fact, been sharply criticized — in modern times especially by
Oscar Hertwig in his work Das Werden der Organismen, the subject of which
is indicated by the subtitle: Eine Widerlegiing von Danvins Zujallstheorie. A
similar judgment was passed by Radl, who, moreover, points out that Dar-
win really applied the social conception of contemporary liberalism to life
in nature; which, as a matter of fact, is at once realized from the acknowl-
edged part played by Malthus's social doctrine in the working-out of Dar-
win's theory. This human-social conception of nature stands out clearly in
the above-mentioned statement regarding the bear, which, if the chance
offers, can take to swimming and develop into a whale. More applicable to
human-social life than to nature is also the form that his utterances often
take of fancies thrown out at random, which reminds one of a social re-
former's improvement schemes rather than of the binding conclusions of a
scientific investigator; "it would be easy," "it would offer no difficulty to
suppose," and other similar expressions frequently occur. This, of course, is
also due to his oft-recurring tendency to allow his thoughts to dally with all
kinds of possibilities — a tendency which, when combined with a belief in
MODERN BIOLOGY 471
the ability of his theory, once advanced, to explain practically any biologi-
cal phenomena whatsoever, is bound to lead to far-reaching conclusions.
Another result of the selection theory is the constant reference to the greatest
possible adaptability to prevailing conditions, with the consequent insist-
ence upon the finality existing in nature. It has already been pointed out how
unsatisfactory this explanation of nature is and further reference will be
made to it later on; here it need only be observed that this belief in a pur-
poseful adaptability to prevailing conditions has in no small degree contrib-
uted towards retarding the development of biology into an exact science.
The influence that Darwin's Origin of Species exercised will be described
in the following. His fame in no wise induced the author to rest on his
laurels; on the contrary, he laboured indefatigably throughout his life still
further to develop the theory that he had created and to apply it to different
life-phenomena. The greatest and most important of his subsequent works
was published in 1868 in two volumes and bore the title Animals and Plants
under Domestication. In the first volume he gives a detailed account of his
intensive racial-biological studies of domestic animals and cultivated plants.
The systematic biology of preceding ages had, as a general rule, depreciated
these beings: they were not true species, only a medley of varieties that no
one could make anything of. Darwin then showed how great is the interest
that this racial research possesses and what important results can be pro-
duced from it. All later racial research is, in fact, based on his initiative. In
point of exactness these investigations of Darwin's certainly cannot be com-
pared with those carried out concurrently by Mendel, but they are far more
many-sided, as regards both material and conclusions, and they also caused
an immense sensation, especially amongst those who led a practical life.
Darwin himself largely had recourse to data provided by animal-breeders
and gardening experts, and he was certainly not very particular about weed-
ing out their alleged results. In the second part of this work he makes fresh
contributions to his descent theory. Here he dilates at length upon his con-
ception of heredity, which played such a radical part in the cultural history
of the nineteenth century, although it is now entirely abandoned. As has
often been pointed out, heredity is to him equivalent to the direct trans-
mission of qualities from the parents to the offspring, a transmission that is
influenced by a vast number of external circumstances. Further, he charac-
terizes atavism — the recurrence of qualities similar to those of earlier
generations — as due to the contrast between the transmission of qualities
and evolution, and, moreover, he points out a number of other hereditary
phenomena — the transmission of qualities confined to only one sex, and the
inheritance of qualities that come out at some special period in life. He also
sought to explain that phenomenon which is now termed the "dominance"
of certain qualities; he calls it the "prepotency of transmission" and finds
472. THE HISTORY OF BIOLOGY
its existence extremely hard to explain, but puts it most closely in connexion
with the age of the qualities in question. Also "latent," or, as they are called
nowadays, "recessive" qualities, he made the subjects of observation and
speculation. Particular care was devoted by him to the problem of hybridi-
zation; he is all the time procuring from his experiments proofs of the over-
lapping of varieties and species. Further, he investigated the cross and
self-fertilization of plants, which he was to deal with more fully in a subse-
quent special work. His speculations on fertilization and hybridization should
not be judged by modern standards; he knew as little as his contemporaries of
the true course of fertilization and so easily became deeply involved in specu-
lations as to the consequences of the effect upon the egg of adequate or inade-
quate quantities of sperm. He then discusses his favourite theory of the laws
of variation, which he now considerably expands, with an increasing ten-
dency towards Lamarckism, external circumstances — climate, food, and
even the use and non-use of organs — being definitely stated as influencing
the forms of variation. Even hybridization and atavism are cited as causes of
variation, besides which the phenomena of correlation are more closely
analysed in connexion with variability.
Pangenesis
The anxiety to find a universally applicable explanation of the phenomena
of heredity and variation led Darwin to think out what he called a "pro-
visional hypothesis of pangenesis." In this theory he gives to the cytology
of the time, with which he otherwise had had nothing to do, a new and
curious interpretation. He believes that every cell, every tissue- or organ-unit
in the body, produces and gives off minute "atoms," which he calls gem-
mules, and that these latter, scattered throughout the body by the currents
of blood and other juices, conjoin as required, and then re-create those
"units" from which they are derived. The sexual products thus contain
' 'gemmules" from all parts of the body, and these are combined in the embryo,
and it is for that reason that all the latter's parts resemble those of the father
or mother, according to whose gemmules have constructed the part of the
body in question. Unused gemmules may be transmitted to the next genera-
tion, with the result that some individuals resemble their father's or mother's
parents. In the same way the bud of a plant is formed by the gemmules of
those parts that are evolved out of it, and the regeneration of the severed
foot of a salamander takes place through the extremity gemmules accumu-
lating at the mutilated end; if, as sometimes occurs, a malformation takes
place, then the wrong gemmules have come into operation. This theory has
been shattered by modern research in the sphere of heredity and need not
therefore be discussed any further in this place; Darwin himself, it is true,
considered it to be only provisional, but he holds that it explains the prob-
lems at issue better than any other theory and should therefore be allowed
MODERN BIOLOGY 47:^
to Stand. This is characteristic of him; the more a theory takes it upon itself
to explain, the more convincing does he consider it to be.^ But exact and
critical research has not dealt thus with the theories; it has set up theories
according as special research has required them, but it has never expanded
them beyond the bounds of absolute necessity. Darwin is here, as so often
elsewhere, a speculative natural philosopher, not a natural scientist.
Darwin on the descent of man
This speculative characteristic is still more conspicuous in his next work,
The Descent of Man, and Selection in Relation to Sex, which was published three
years after the former book, but which was likewise written after many
years of preparation. In The Origin of Species he had already in passing ex-
pressed the opinion that natural selection would without doubt eventually
throw light also on the origin of man — an assertion that was enough to
excite very great attention. The subject had already been taken up by others:
by Huxley and, above all, by Haeckel, and it was thus no longer a matter
of real urgency. Darwin's presentation of it, however, possesses an interest of
its own. His arguments that man has through natural selection by means of
the struggle for existence been evolved from a series of animal forms are,
of course, the same as those he had previously developed in regard to the
animals; the anatomical and embryological argumentation he was able to
borrow from his above-mentioned predecessors. It may be pointed out,
however, that he does not insist upon man's relationship with the anthropoid
apes, as Haeckel has done; he observes, it is true, physical and psychical
agreements, but otherwise maintains for the most part man's character of a
mammal. Of greater interest, however, is his derivation of the human
psychical qualities; he analyses a number of such qualities of different kinds —
curiosity, the tendency to imitate, memory, imagination, reflection — and
he finds them existing also in the animals. He even notices an equivalent to
religious feelings in the dog's awe of his master. On the whole, he falls into
the same error as innumerable animal psychologists since then, of letting
qualities that man has through training inculcated into his domestic animals
be regarded as spontaneous manifestations of the intellect. As to the existence
of moral qualities, he refers to the characteristics of self-sacrifice and social
sense to be found in many animal forms — in regard to the ants he holds in
this respect the same exaggerated ideas as many of his contemporaries —
^ As an instance of how boldly Darwin takes up the most difficult problems for discussion,
and how casually he afterwards solves them, the following may be cited (Variations, I, p. 8).
He maintains that, in spite of natural selection, very simple life-forms have nevertheless been
preserved through the ages by adapting themselves to very simple conditions of life: "for what
would it profit an Infusorial animalcule for instance or an intestinal worm to become highly
organized?" It must be admitted that, if the problem is difficult to solve, the answer certainly
makes it none the easier.
474 THE HISTORY OF BIOLOGY
and maintains that even amongst wild tribes only social virtues are respected.
He has no interest in individual soul-life — a lack of interest which he like-
wise shared with many scientists of his age and which involves him in
anthropomorphitic interpretations of purely instinctive phenomena, not to
mention the credulity that he shows towards the statements of other owners
of domestic animals regarding the purely human intelligence manifested by
their four-legged friends. As to the time and place of the first appearance of
the human race he expresses himself with a certain amount of caution, as he
does also in regard to the racial problem.
Sexual selection
By far the greater portion of the work under discussion deals, however,
with another question — namely, the origin of the secondary sexual char-
acters. To these Darwin considers that the theory of natural selection
in the ordinary sense cannot be applied; he does not believe he can use it
to explain the origin of such features as the horns of the stag-beetle and
the males of other coleopters, the brilliant coloration of male butterflies,
the cock's-comb, the horns of the stag, and other similar characteristics.
He considers rather that these features have arisen as a result of special
sexual selection; the males have competed for the favour of the females,
and the most attractive or the strongest have gone off victorious and been
allowed to propagate and to transmit their characteristics by inheritance
to their offspring. He finds proofs of this in the playing and the fighting
that takes place between the males in the mating-season; the butterflies'
sport in the air, the combats of cocks and stags, the song of the nightingale
and the lark, the play of the wood-grouse, and the stately mating-dance of
the cock of the rock. But it is not only the male qualities, but also certain
common characteristics that he attributes to this kind of selection, as for
instance the coloration of the butterflies, which he believes to have arisen
owing to the females' also having acquired their share of the inheritance
of sexual selection. This doctrine of sexual selection was rejected even earlier
than the general theory of selection and is nowadays embraced by hardly any
true scientists, although popular literature shows traces of it. What really
brought about its rejection is the increased knowledge of internal secretion
and the connexion of the secondary sexual characters with it; both sexual
coloration and mating-play have their explanation in this. That Darwin
knew nothing of this cannot, of course, be laid at his door, but even apart
from this fact, the sexual-selection theory certainly gives strong evidence of
his tendency to attribute without criticism purely human ideas to the ani-
mal kingdom, to believe in "beauty competitions" among butterflies and
beetles, fishes and newts, or that grasshoppers and crickets have a musical
ear. It has also been pointed out that it is purely physical strength and not
beauty at all that makes cocks and stags successful with the females,
MODERN BIOLOGY 475
besides which it may often happen that the strongest males spur or butt one
another to death, with the result that afterwards comparatively weak speci-
mens win a place among the females. In support of his theory Darwin placed
the male intelligence at a radically higher value than the female. He over-
looked the fact that the females also exercise an important function, which
likewise demands intelligence, in the care and protection of their offspring.
His theories on this subject nevertheless won strong support in certain liter-
ary quarters; it is well known that, among others, Strindberg has referred
to them with enthusiasm.
In connexion with the last-mentioned work Darwin published another
book, entitled Expression of the Emotions in Man and Animals, in which he
records a large number of facts regarding emotions in man and the animals,
which he had amassed and compiled and to which he, of course, applies
his theory of selection and descent. Further, in his later years he published
a number of works on special subjects that are in part extremely valuable.
Among these may be mentioned his work on insectivorous plants — it was
he who first pointed out that these plants really digest and resorb the im-
prisoned animals — another on the climbing organs of plants, in which these
organs are described with exhaustive thoroughness and from numerous fresh
points of view, and finally a work on cross- and self-fertilization in plants,
as also a book of fundamental importance wherein he continues Sprengel's
work, which he had rescued from oblivion, and paves the way for modern
heredity-research. In the year before his death he also published a brief but
ingenious work on the formation of vegetable mould through the action of
worms, in which he establishes, on the strength of a large number of ob-
servations and experiments, the important role played by these animals as
re-formers of the earth's surface, in that a considerable portion of the earth's
outer layers passes through their intestinal canal and is thereby influenced
physically and chemically — facts which research had previously failed to
observe, but which have latterly been fully confirmed.
Darwin s general opinion of life
During the greater part of his life Darwin devoted himself to his own par-
ticular field of research more thoroughly than most other scientists. He never
went in for teaching nor took up any other public appointment, while owing
to his ill health he had to give up social life, with the result that his activi-
ties became more and more confined to biological speculations and experi-
ments. This may explain why he embraced with such intensity, but also
with such limitations, the theories he set up. He was but little influenced
by other natural-scientific tendencies, eventually losing interest even in gen-
eral cultural problems. In his youth he had been interested in art, poetry, and
music, but in his old age even these lost their attraction for him. True, by
way of diversion he used to have novels read to him, requiring only that
476 THE HISTORY OF BIOLOGY
they should have a happy ending; he paid but little attention to literary
faults. And his religious interests went the same way as the literary; the
Christian faith of his youth had undoubtedly been traditional from the very
beginning, without any feelings of personal experience; his faith died grad-
ually and without any crisis, leaving behind a peaceful and untroubled res-
ignation in face of the ultimate problems of existence, a resignation which
was never disturbed by anything except the innumerable senseless and irra-
tional inquiries he received on the subject and to which he invariably replied
conscientiously. It is worth quoting the following out of one of these re-
plies as a final touch to the description of his character: "The safest con-
clusion seems to me that the whole subject is beyond the scope of man's
intellect; but man can do his duty."
Judgments on Darwin
Very different judgments have been passed on Darwin. Even on his first
appearance he was either extolled as one of the greatest geniuses in the
world or abused as an ignorant and unreliable dilettante, according to the
different points of view. Nor have subsequent generations been any more
unanimous; especially since the theory of selection has been condemned —
at least in its original form — hard words have not been spared against its
creator — as a matter of fact, a natural reaction against the adoration meted
out to him towards the close of his life, which received confirmation in his
being buried beside the grave of Newton. Was this apotheosis justified or
not? This question has been answered and can still be answered either way.
To raise the theory of selection, as has often been done, to the rank of a
"natural law" comparable in value with the law of gravity established by
Newton is, of course, quite irrational, as time has already shown; Darwin's
theory of the origin of species was long ago abandoned. Other facts estab-
lished by Darwin are all of second-rate value. But if we measure him by his
influence on the general cultural development of humanity, then the prox-
imity of his grave to Newton's is fully justified. It is certain that since the
days of the latter no scientist has so deeply influenced man's general concep-
tion of life as Darwin has done; it is his theory of evolution that has taken
the place of the idealistic theory of romanticism and made the common de-
scent the connecting link in existence instead of ideas and archetypes. In
all spheres of knowledge the development from earlier to later stages has
been the one clue for research; history, which had previously sought for
"guiding ideas," is now an evolutionary science, just as is philology, and
even philosophy has at least one school that has followed the same prin-
ciple. Everyone knows the important role played by the idea of evolution
in naturalistic literature. The influence of Darwinism on biology will be
described in the next chapter. Of its weaknesses a certain number have al-
ready been referred to above; it shared with all new ideas the illusion that
it could do too much; this was so with Darwin himself, modest though he
personally was, and still more so with his admiring successors. We shall
now proceed to describe the differences of opinion caused by the new doctrine.
CHAPTER XII
FOR AND AGAINST DARWIN
M
Why Dartvin s theory prevailed
'oDERN CRITICS have often asked themselves how it is that a hypothe-
sis like Darwin's, based on such weak foundations, could all at
once win over to its side the greater part of contemporary scien-
tific opinion. If the defenders of the theory refer with this end in view to
its intrinsic value, it may be answered that the theory has long ago been
rejected in its most vital points by subsequent research. It has also been
pointed out, for instance by Radl, that the objections made against the the-
ory on its first appearance very largely agree with those which far later
brought about its fail. The factors governing the victory of Darwinism thus
represent a problem of the greatest importance, not only in the history of
biology, but also in that of culture in general — a problem that would re-
quire far more exhaustive treatment than can be given to it here. In this
work we can only endeavour to throw light on some of the circumstances
that appear to be specially remarkable surrounding this important episode,
the history of which it will largely be the duty of future generations to
write.
Darwinism and liberalism
Darwin's origin of species contains many points that were likely both to
win the applause of and to give ofTence to his contemporaries. A factor that
without doubt very largely contributed to both the one and the other was
the book's relation to the political movement of the time, to which refer-
ence has already been made. From the beginning Darwin's theory was an
obvious ally to liberalism; it was at once a means of elevating the doctrine
of free competition, which had been one of the most vital corner-stones of
the movement of progress, to the rank of a natural law, and similarly the
leading principle of liberalism, progress, was confirmed by the new theory —
the deeper down the origin of human culture was placed, the higher were
the hopes that could be entertained for its future possibilities. It was no
wonder, then, that the liberal-minded were enthusiastic; Darwinism must
be true, nothing else was possible. But beside this there was a good deal
more in it that could attract radical cultured views, chiefly its strongly
worded polemic against the doctrine of creation, which could be employed
to counteract theological obscurantism, and also the very idea of a material
477
478 THE HISTORY OF BIOLOGY
connexion in existence, a principle that could be set up in opposition to
the theories of ideas held by reactionary romanticism. The deficiencies in
Darwin's work were therefore readily overlooked — his vague starting-
point, his uncritical material, his weak arguments based on loose assump-
tions, his belief in the power of chance and of finality as an explanation of
nature. As a matter of fact, the natural explanations of the preceding ages
failed still more in that respect; they were generally based on the wisdom
of the Creator and the benefit of man as the cause of all that exists and takes
place — that is to say, an explanation without the slightest trace of scien-
tific treatment. Darwin's theory, then, was at any rate an immense advance;
its weaknesses could be overcome by continued research, its vagueness and
casualness removed by fresh discoveries and replaced by firmly established
facts, while the finality in nature could thus be made synonymous with
natural law. Briefly, no one was prepared to doubt the possibilities of the
theory's future development, and for the moment it entailed a freedom from
the pressure of prejudice which there had previously seemed to be no means
of avoiding.
Defiance of the conservatives
While, then, liberal tendencies felt themselves closely bound up in Dar-
winism, the new movement was for that very reason all the more repugnant
to the conservative social elements. Those who looked for their ideal in
the past and in tradition must have been appalled to see the good old times
depicted as a kind of half-way station along the road from the ape stage;
and that free competition which to their mind only led to all manner of li-
cence, was that to be the true creator of the life that is lived today, instead
of the divine reason which has governed the world and preserved law and
justice? And, again, this vague, indeterminate idea of evolution, was it to
be substituted for those firmly established and eternal ideas that governed
the creation of nature and its forms? Thus reasoned many, and Darwin's
theory was therefore challenged from pulpit and professorial chair, at sci-
entific gatherings, in journals and newspapers. This first polemic against
Darwinism has its own peculiar interest; it is dazed and not particularly
keen-sighted, it clings despairingly to the old ideas and as yet lacks orienta-
tion as to the exact position adopted by its opponents. In the present history
it is only possible to give attention to some few of the more representative
scientific contributions, whereas the miscellaneous mass of protests from
other quarters can have no place here.
Owen s opposition
The highest scientific reputation among the opponents of Darwin was un-
doubtedly that of Richard Owen. It was, of course, impossible for the lat-
ter's idealistic morphology to be reconciled with the Darwinian doctrines
of descent, and if anyone was to discover and demonstrate the weaknesses
MODERN BIOLOGY 479
underlying the new theory, it was he. Tlie fact, however, that his influence
was not so great as his scientific reputation might have warranted was mostly
due to the way in which he conducted himself; instead of openly defending
his views he wrote anonymously, repeatedly referring to "Professor Owen"
as his authority in opposition to Darwin. This gave his contribution a tinge
of lampoonery that detracted from the effect it might otherwise have had.
The article (it appeared in the Edinburgh Kevieiv of i860), which much em-
bittered Darwin, is chiefly interesting as being an expression for the sharp
contrast between the romantic natural philosophy and the realistic evolu-
tional theory. Owen points out with strong emphasis how few are the facts
and how weak the proofs that form the basis for the new theory, how the
problem of species-formation must still be considered unsolved in spite of
the theory of selection, how it was possible to assume other factors govern-
ing species-formation besides variation and natural selection. As such fac-
tors he suggests parthenogenesis and alternation of generations; he believes
it possible to suppose that the various stages in such a cycle — polypus and
medusa, or sporocyst, redia, cercaria — might, so to speak, liberate them-
selves from the series and begin to give rise to forms similar to them-
selves, with the result that the whole cycle would disintegrate into a number
of widely differing life-forms. He even adopts Pouchet's spontaneous-gener-
ation experiments in his support against Darwin: if the Infusoria spontan-
eously generate daily, how can it be assumed that all higher beings could
have been evolved in one single series originating in primitive forms? Owen's
suggestions in regard to species-formation are certainly not very happily con-
ceived from a modern point of view, and indeed they are only presented as
experiments with ideas in order to prove how complicated and difficult of
solution the problem of species-formation really is, but the worst of it is
that Owen brings into the field the whole of the thought-systems of the old
idealistic natural philosophy; as a factor that actively operates in the crea-
tion of the symmetrical forms of the higher animals he adduces a "polar-
izing force," the true essence of which need not be analysed here, as the
name itself explains it. Even the old doctrine of " the ideal type" is brought
forward for the same purpose. But one who has recourse to such empty
phrases to explain the origin of life-forms has no right to accuse Darwinism
of weak argumentation and of making false hypotheses. A controversy such
as this best shows what an immense advance Darwinism nevertheless in-
volved at the time, and at the same time explains why it is that even the
authorized objections of the old school must die away unheard.
One gets the same impression from the criticism of Darwinism offered
by Agassiz, another important representative of the old biological school.
Jean Louis Rodolphe Agassiz was born in 1807 at Motier in Switzerland, of
French parents, and even during his school-time devoted himself to natural
480 THE HISTORY OF BIOLOGY
science. He studied at several German universities, his teachers including
both Schelling and Oken, but principally Dollinger, mentioned above as
von Baer's master; he became doctor of medicine and afterwards spent a
couple of years in Paris in lively discussion with both Cuvier and Humboldt.
His chief object of study was the fishes, both recent and fossil; a large work
that he had commenced on the fishes of Europe was never finished, while
another on fossil fishes proved a pioneering work in its own sphere. But
besides this, glacial research proved of special interest to this many-sided
scientist, and in this field too he was a pioneer. He proved that the glaciers
had in earlier times been far more extensive in his native country than they
are now, and during a journey to Scotland he found that large glaciers had ex-
isted there too in past ages. From this he drew conclusions regarding the
general glacialization of Europe, which afterwards led to that highly de-
veloped research-work on the glacial period which has been especially note-
worthy in Scandinavia. During the years 1831-46 Agassiz was a professor
at Neuchatel; he then moved to America and became professor at Harvard
University. There he did splendid work as both a zoologist and a geologist,
making extensive journeys and producing works on the animal world and
the zoology of America, as well as on theoretical problems. He died in 1873.
In his theoretical writings Agassiz shows himself a true romantic natu-
ral philosopher, as might be expected from the education he received. The
problem of species engaged him a great deal and is solved by him in a mark-
edly idealistic direction. In such circumstances it was obvious that he could
not hail the advent of Darwinism with any great enthusiasm. In his polemics
against it he makes a great point of its weaknesses; lack of observation of
the real transition from one species to another, lack of obedience to law in
its theory of natural selection, the weak conclusions drawn from similarity
in the embryonic stage to similarity of origin. But the most serious mistake
to his mind is that the new theory fails to realize the creative idea running
through all animate nature. The individuals perish, but hand over to their
posterity, generation by generation, all that is typical, with the exclusion of
what is merely individual; therefore, while the individuals have only a mate-
rial existence, species, genera, families, and so on upwards exist as the
thought-categories of the Supreme Intelligence, and as such possess a truly
independent and immutable existence.^ Here, it will be seen, speaks the pure
romantic idealism, whose supporters, thanks to their intensive professional
insight, have no difficulty in discovering the weaknesses underlying the new
biological theory, though only to maintain in its stead their own
^ In his "Essay on Classification" Agassiz, speaking of rudimentary organs, maintains
that these exist not for any purpose of function, but to complete the design, just as in a building
certain details are introduced for the sake of symmetry, without any idea of their serving a
practical purpose.
MODERN BIOLOGY 481
Speculation, equally unworkable in form as in contents and therefore in-
evitably doomed to failure.
It would hardly be worth while to carry this account of the attacks
against Darwin any further. We might still mention the contribution of
S. WiLBERFORCE, Bishop of Oxfotd, owing to the sensation it created at the
time. Having himself studied natural science, and with the indefatigable
Owen as prompter, he reviews the weaknesses of Darwinism in an easy and
fluent style, though somewhat superficially, but at the end of his treatise
he spoils his case completely by sermonizing on the subject of the origin
of man, bringing forward all the persons of the Trinity as arguments to
prove a special creation in the image of God. From such opponents Darwin
clearly had nothing to fear. But even scientists with a truly modern concep-
tion adopted from the outset an attitude of criticism towards this theory —
KoLLiKER, for instance. In a brief examination, substantiated by numerous
facts, Kolliker submits in a concise and determined style his objections to
the theory of selection, at the same time acknowledging the great service
of Darwin in having sought to base the knowledge of the origin of organ-
isms upon experiments and in having made descent the foundation thereof,
so that the life-forms might be regarded as a series of evolutionary phenom-
ena. He expressly declares that the earlier attempts of natural philosophy to
construct a history of evolution are weak in comparison with Darwin's,
and, moreover, he appreciates the far-reaching insight and the splendid con-
scientiousness on which his theory is founded. As its weak points he men-
tions first of all its teleological conception; the principle of finality as applied
to life -forms, which has already been pointed out above; further, the ab-
sence of transition forms between the species, both extant and fossil, the
lack of proof that characterizes 'the entire hypothesis of selection, and
finally the circumstance that nothing is known of unfertile variety-hybrids,
which would nevertheless be bound to appear somewhere if the varieties
were transitions to species. Moreover, Kolliker holds that it is possible to
imagine other ways of evolution than Darwin's. He considers that the idea
that all species have been created as they are is not worth discussing, but
it is conceivable either that all organisms have arisen each out of its own
primary form, or that the species have come into existence through one pri-
mary form or through a few. The latter alternative he considers to be more
probable, but then there must be a common law governing formation, ac-
cording to which forms of one kind may in certain circumstances give rise
to entirely different forms, either by a larval form's adopting an independ-
ent course of development, or by an egg or embryo of a lower form's giving
rise to a higher type of life. This creation theory of Kolliker's is merely a
concept and is, moreover, based on hypotheses that have never been con-
firmed. Of real value, on the other hand, is his criticism of Darwm's theory
48x THE HISTORY OF BIOLOGY
which is founded on a truly exact, and not on a natural-philosophical,
basis.
Huxley versus Kolliker
This criticism of Kolliker's was opposed by Huxley, who vehemently de-
nies that there is any teleological explanation at all in Darwin, whose en-
tire theory is based rather on the absence of any creative purpose in nature.
And in proof of his view Huxley cites exactly the same quotation out of
the Origin of Species as Kolliker does for his own argument. From this it is
obvious that the two antagonists must be standing in some essential respect
on different ground, and the question is of such great general interest that
it deserves closer examination. Strictly speaking, Huxley is right, in so far
as no creative design in the romantic natural-philosophical sense is ever re-
ferred to by Darwin; but this does not prevent his constant assertion as to
the adaptability of life-forms and organs to certain given conditions from
implying a teleological explanation of phenomena; not only the entire the-
ory of sexual selection, but also most of the doctrine of natural selection
actually rests on this assumption. The contrast between the romantic and
the Darwinian teleology is best explained by an example. It is asked: Why
has a cat claws? For the sake of the creative design, say the romanticists,
and in order to serve the purposes of the cosmic order. For its own sake,
says Darwin, and in order to enable it to survive in the struggle for exist-
ence. But it is really the question itself that is absurd — as absurd as the
question: Why does a stone fall? or Why does the earth revolve round the
sun? Biology can only endeavour to find out the conditions under which
cat's claws are developed and used, but never anything more; those who
question beyond that fail to fulfil Bacon's requirement that we should "ask
nature fair questions." But Darwin and his contemporaries are constantly
putting such wrong questions to nature. This is, of course, due to the fact
that they were unable to free themselves entirely from the influence of ro-
mantic philosophy, which, indeed, they desired to abandon and the weak-
nesses of which they fully realized, but its grasp of the problem of life was
really too firm for them to loosen. Natural philosophy had, indeed, found
in its plan of creation an explanation for everything, and to resign in face
of the causes of the phenomena of life would have meant, to the new direc-
tion in which biology was moving, almost the same thing as a declaration
of bankruptcy in face of its opponents. And in contrast to the idealistic plan
of creation Darwin's teleology involves possibilities of development, in so
far as a number of the so-called purposeful adaptations have since, mainly
through modern researches into the problem of heredity, found its law-bound
explanation, while other phenomena have had to accept that resignation in
face of the inexplicable, which is the hall-mark of exact and critical science.
Darwin's theory of adaptation, which is now so often condemned for its
MODERN BIOLOGY 483
credulity, has thus in reality formed a necessary transitional stage, which
has freed biology from the illusions of the past and made a more exact re-
search possible in the future.
Among Darwin's other opponents in Germany may be mentioned Al-
bert WiGAND (i8ii-86), professor of botany at Marburg, a pupil of Schlei-
den, and well known as a capable plant-anatomist and plant-physiologist,
as well as a leading expert on cryptogams. He was, moreover, deeply re-
ligious and on that account was unable to accept the theory of spontaneous
generation. It was therefore inevitable that Darwinism should have been
odious to him from the start, and he wrote many treatises against it. Even-
tually, after ten years of preliminary work, he summarized his views in a
work comprising nearly thirteen hundred pages, entitled Beitrdge :^ur Metho-
dik der Naturforschung. He here shows himself to be a keen-sighted student
of nature and a keen critic of the old exact school. Cuvier is his ideal as a
scientist and he definitely associates himself with him in his opposition to
Geoffrey's efforts to attain natural-philosophical unity. He has a keen eye
for the weaknesses of Darwinism and analyses them objectively and in de-
tail; he especially brings out the weaknesses underlying the theory of se-
lection, and in contrast to the lack of design in the phenomena of variation
and selection, as presented by Darwin, he maintains the existence of a defi-
nite course and plan in evolution — a plan that excludes both chance and
explanations of finality. This criticism is, indeed, on the whole justified,
and even Wigand's assertion that Darwinism is natural philosophy rather
than exact research is quite a fair judgment; but when it comes to trying
to justify the idea of conformity to law urged in opposition to the doctrine
of chance, the former is ascribed to a personal deity, for natural science can-
not get away from an ultimate cause of existence. This, of course, should
not be used as grounds for a natural-scientific explanation, but the doctrine
of the creation and the theory of the immutability of the species, which
Wigand would urge in opposition to Darwinism, are nevertheless based
upon it. In doing so, however, he has vitiated the effects of his criticism;
his ideas were capable of satisfying neither the natural philosophers of the
old school nor the exact scientists, and he himself lived just long
enough to see Darwinism reach the height of its influence upon human
culture.
Even the aged von Baer entered the lists against Darwinism, complain-
ing of its lack of conformity to natural law; he sees in evolution a striving
after a definite goal — " Zielstrebigkeit,'' as he calls it — and this, indeed,
explains the finality in existence, but presupposes in its turn a common
scheme for all natural phenomena, which is only conceivable with a per-
sonal creator as the primary cause. The high respect in which this octoge-
narian student of evolution was held, exempted him from harsh criticism
484 THE HISTORY OF BIOLOGY
at the hands of the younger generation; his contribution was added to the
records in silence.
Darwinism was least appreciated in France, where Cuvier's pupils held
sway in the realm of zoology and where even representatives of experimen-
tal research — Bernard and others — had little sympathy for the specula-
tive and hypothetical elements in the new theory. It is striking that Darwin
was not elected to the French Academy of Science until after he had pub-
lished his works on plant-physiology, and then under reference only to them
and not to the descent theory. And when this theory — or " trans fortnisme,' '
as it was called in French — eventually found acceptance in the country of
Lamarck, it was with him rather than with Darwin that the followers of
the new tendency associated themselves. Of the earlier critics of Darwinism
in France the first name is that of Jean Louis Armand de Quatrefages de
Breau (i8io-9x), first a physician and finally professor of anthropology at
Paris, and famous as a leading specialist on marine fauna, particularly the
Annelida, but foremost as an anthropologist. As this last he carried out
valuable investigations into special subjects, all, however, governed by a
firm conviction as to the unity of the human race and its independence of
other life -forms. He wrote a number of treatises against Darwinism, the
chief of which was one entitled Charles Darwin et ses prkurseurs fran^ais, in
which he begins by describing several transformistic authors of French na-
tionality: de Maillet, Buffon, Lamarck, and others. In regard to Darwin,
Quatrefages admits that there is a struggle for existence, but does not be-
lieve in its power to create new life-forms. He sharply criticizes Darwin's
habit of adducing the probable and the possible — purely personal convic-
tion instead of facts proved on conclusive evidence — and he particularly
points out that, when it comes to the question of the life-phenomena of
past ages, Darwin constantly appeals to "the unknown." And Quatrefages
concludes his critical examination with the words: "Let us not dream of
what may be; let us instead assume and seek what is!" Among the earlier
critics of Darwinism Quatrefages is worthy of respect on account of the con-
siderate and objective manner in which he passed judgment on the theory.
Eventually, however, the descent theory gained ground even in France,
chiefly, as mentioned above, in the Lamarckian form, which at the same time
became known in other countries also, and which will be described later on.
Vast quantites of polemical writings against Darwin and his theory
appeared during the period immediately after he first attracted public at-
tention; most of these were of practically no scientific value, since they were
based on religious arguments, which were the most usual, or else on quasi-
scientific or other grounds. Of the really objective contributions to the sub-
ject it would be possible to name many others besides those referred to above,
but space forbids a more detailed review of them. At the same time there
MODERN BIOLOGY 485
came forward in defence of Darwinism many distinguished scientists, who
made weighty contributions to the discussion and assisted in the rapid ad-
vance along the new lines laid down by Darwin. Even of these it is possible
only to name a few; in the present chapter reference will be made to the
English contributions in favour of Darwin, while one or two separate chap-
ters will be devoted to the development of Darwinism in Germany, where
it acquired an essentially novel character.
Darwin's supporters
Among the first to associate themselves with Darwin was the aged Lyell.
In a work entitled Geological Evidence ofi the Antiquity of Man, published in
1863, he takes up the question of the origin of species by means of varia-
tion. He refers briefly to Darwin's theory and in support thereof cites a num-
ber of facts, especially geological and palieontological; of these he bases
his argument mainly on the extinct proboscideans of the Tertiary period,
while he further adduces a number of fossil insects, as well as the saurian
bird Archasopteryx. In regard to man, whose primitive history had been
the real subject of the book, sympathetic reference is made to the statements
of, inter alia, Huxley, as to man's anatomical agreement with the higher
apes; similarly, mention is made of Darwin's theory of the origin of the
intelligence by means of natural selection, and the work concludes with a
refutation of the accusation that Darwinism would lead to materialism.
Darwin himself highly appreciated the support thus given him by Lyell,
and the influence of the aged geologist certainly contributed much towards
bringing the new doctrine to victory.
Among those who, besides Darwin, should be named as supporters of
the theory of selection, the first place is due to Alfred Russell Wallace.
Born in 1813, he was originally an engineer and afterwards a schoolmaster,
and he was besides interested in collecting plants and insects. In 1848, in
company with his friend Henry Walter Bates (i82.5-9z), he made a voy-
age to Brazil for the purpose of exploration and the collection of scientific
material. After a year the two friends parted; Bates remained in Brazil,
while Wallace returned home and shortly afterwards made a journey to the
East Indian archipelago, where he remained for a number of years, continu-
ing his comparative biological studies on the various islands. Upon his re-
turn home he found himself already a famous man and continued his bio-
logical research-work, partly on voyages and partly in his own country.
He never received any permanent appointment, but had to earn a living as
a writer and lecturer. In his old age he was assured a means of subsistence
through a government pension. He died in 191 3, over ninety years old.
Wallace' s discoveries in animal geography
The result of Wallace's Indian journey proved to be of the greatest impor-
tance; he thereby became one of the pioneers of modern animal-geography.
486 THE HISTORY OF BIOLOGY
He established the fact that the western half of the archipelago possesses
an essentially Indian animal world, whereas the eastern half has an equally-
marked Australian fauna; the border-line he found to lie in the narrow but
deep sound between the islands Bali and Lombok, and northwards from there
in the Macassar Strait between Borneo and the Celebes. He afterwards com-
piled, with the aid of the results gained during this and subsequent voyages,
an animal geographical system, in which the globe was divided into sep-
arate regions based on the distribution of animal forms both in recent times
and in preceding periods. This animal geographical system, which is univer-
sally known from the text-books on zoology, is a contribution of lasting
value to the development of biology.
But in the course of his studies of the distribution of animal life in the
East Indian islands Wallace found himself faced with the same problems as
Darwin in the Galapagos Islands; the various islands and island-groups pos-
sess their peculiar animal species. The distribution of species on the earth is
thus governed by geological conditions, and if we consider the animal life
of earlier periods we find that, instead of the new extant forms, there were
other similar forms — in fact, that, as he says, every species has been pre-
ceded in time and space by a similar species. These reflections he recorded
in a treatise which he sent home and which was printed in 1855. The ex-
planation of the phenomenon he found — like Darwin — when meditating
upon Malthus's theory of competition; it is the struggle for existence that
has compelled living creatures to develop themselves in order not to perish
in the struggle against other species; if a variety has been equipped with
more powerful qualities than the main species, it drives out the latter and
usurps its place. This theory Wallace expounded in a report, which he sent
to Darwin for perusal; the latter was struck by the agreement with the
ideas that he himself just happened to be working out and found the situa-
tion highly embarrassing. At the suggestion of some friends he published
Wallace's treatise together with a report of his own results, which he sub-
mitted to the Linnean Society in 1858, thus giving science an opportunity
of seeing the same theory presented by two investigators working independ-
ently. Much surprise has been expressed at the incident, which has often
been put forward as a proof of the undeniable truth of the theory. It is pos-
sible to find an explanation of the phenomenon by making a comparison
between the two originators of the theory; they were both self-taught men
with essentially systematic interests, but without any anatomical training;
they had both explored an island region and received their impressions there-
from; both had consequently been confronted with the problem of the dis-
tribution of species, and, finally, both had been influenced by Malthus; and
though the fundamental view-point is the same in both scientists, yet Wal-
lace has a conception of the problem that is in many respects peculiar to
MODERN BIOLOGY 487
himself. He manifestly never felt so deeply moved by the actual doctrine of
the creation as Darwin had been, and, further, he has by no means the same
interest as Darwin in domestic-animal varieties, with which he himself had
never experimented; rather, having studied in the richest tropical regions,
he had gained a far stronger impression of the wealth of life-forms and their
adaptation to environment. This especially comes out in the mimicry theory
that he and his friend Bates created.
Theory of protective resemblance
After a ten years' sojourn in the tropics of South America Bates returned
home with rich collections and eventually became secretary to the British
Geographical Society. He wrote an essay in which he propounded the idea
of protective resemblance in the animal kingdom — an idea that was after-
wards taken up and further developed by Wallace. It is known that a large
number of animals possess external characteristics that correspond to con-
ditions in the natural surroundings in which they live; the white fur of
polar animals, the sandy yellow of desert beasts, the likeness of many in-
sects to the bark of the trees on which they live, are all examples of this. In
the more abundant plant-life of the tropics there appear still more remark-
able instances of this similarity, especially among the insects; well-known
examples are the "wandering leaves" and "wandering sticks," which,
owing to their likeness to the undergrowth, often elude the observation
of even the most experienced collectors. Wallace believes that all these forms
have arisen through the circumstance that natural selection in the struggle
for existence has favoured those individuals that, owing to variations in
the direction of greatest likeness to their surroundings, have been better pro-
tected than others and have thereby had a better chance to propagate. But
Wallace considers that even the obvious exceptions from the rule which quite
often occur — animals with strikingly brilliant colours — only still fur-
ther confirm the law, seeing that they really possess some other character-
istic which acts as a powerful protection against their enemies and which
thus converts their splendid colours into a kind of warning signal to the
latter; as, for instance, an offensive odour, as in the skunk of America, the
natterjack, and the salamander, as well as a large number of insects, in-
cluding our common lady-bird, with its magnificent red-and-black spotted
wings; or, again, a hard shell, as in many brilliantly coloured tortoises; or
poison, as in many of the vividly marked snakes in the tropics. The most
remarkable application of this law Wallace sees, however, in the mimicry
or disguise whereby certain animals protect themselves against their enemies
by resembling other more dangerous animals in their outward appearance;
there are flies that in form and manner of flying resemble bumble-bees, butter-
flies that resemble wasps; and this disguise is demonstrated still more in the
tropics, where non-poisonous snakes are often misleadingly like the poisonous
488 THE HISTORY OF BIOLOGY
ones and many insects exhibit similar [congruities. Even this protective
resemblance Wallace, of course, derives from natural selection. Furthermore,
he points out in this connexion the females' need for protection during the
period when they are ministering to their young as a cause of their less
conspicuous coloration, as in the birds, whereas the cock birds, which do
not require this protection, have developed greater splendour of colouring.
Wallace thus explains the external dissimilarity of the sexes without having
recourse to Darwin's theory of sexual selection, which he rejects.
The whole of this theory of protective resemblance, which was once
cited as one of the strongest arguments in favour of Darwinism, has natu-
rally been discredited concurrently with the theory of selection itself; the
mimicry theory in particular had already been vehemently attacked by sci-
entists who did not find it accord with their observations and experiments;
the enemies that through their pursuit of prey were supposed to have called
for a protective resemblance have in many places been found to be non-
existent, and remarkable instances have been discovered of resemblances of
this kind in animals in different parts of the world, which could not there-
fore have influenced one another's appearance. Further, in order to maintain
the theory it has been necessary to ascribe to a great many animals powers
of observation and distinction as weak as man himself possesses. In the lat-
ter respect Wallace was extremely credulous; he states, inter alia: "The atti-
tudes of some insects may also protect them, as the habit of turning up the
tail by the harmless rove-beetles (Staphylinidas) no doubt leads other ani-
mals besides children to the belief that they can sting." This comparison
between the animal's power of observation in nature and that of a child is
certainly very naive. But Wallace was on the whole more of an imaginative
than critical nature; very soon he had astonished the world by becoming a
convinced spiritualist, although he was a free-thinker in religious questions,
and in later years he became entirely engrossed in spiritual seances and a
number of similar fantastic ideas of spiritual life in nature, while at the same
time he expressed his utter contempt for the results of modern heredity re-
search. He thereby placed himself definitely on the side of natural-scientific
development.
A personality of an entirely different character was Darwin's other cham-
pion in England — Huxley, one of the most famous biologists of his time.
Thomas Henry Huxley was born in 182.5 in a London suburb, the son of a
poor schoolmaster. After two years at school, which he himself described
as "a pandemonium," he had from the age of ten to pursue his studies by
himself and he did so with such success that seven years later he gained an
entry into the medical faculty in London. Having passed his examinations,
he became a surgeon in the English fleet and served in that capacity in a
vessel that was exploring the channels north of Australia. In these tropical
MODERN BIOLOGY 489
waters there existed abundant animal life, which induced the young doctor
to investigate it. Among other works he brought out a book on the medusre,
which brought him wide recognition. At the age of thirty he was appointed
professor at a School of Mines and this led him to take up palasontological
research, but he further had an opportunity of teaching physiology and com-
parative anatomy, which he had already thoroughly studied during his stu-
dent days. Full of energy and initiative as he was, he was able to make
practical use of his science to an extent that few have equalled. Not only
by means of popular lectures and text-books, but also as a member of school
committees and an expert on fishery questions, he laboured to expand the
knowledge of biology and to increase respect for its methods and mode of
thought. His authority ultimately became very great and honours of all
kinds were showered upon him — more, in fact, than he cared for. After
a long period of suffering he died in 1895. His marble statue stands by the
side of Darwin's in the South Kensington Museum in London.
Huxley' s ivork on the medusce
Huxley was a highly gifted scientist, though critical rather than creative.
His first work on the anatomy and affinities of the medusa; was that of
a pioneer; he therein demonstrated the connexion between hydroid polypi 1
and hydromedusa; and combined them into one order, the Hydrozoa. Of
still greater value was his idea of comparing the dermal and intestinal layers
of these animals with the germinal layers in the embryonic stages of the
higher animals; out of this comparison eventually arose the general theory
of germinal layers. His sea voyage likewise produced a series of valuable
studies on the Tunicata. He also made important contributions in the sphere
of vertebrate anatomy; especially well known are his com.parative studies
of the structure of the cranium, whereby he proved, on the support of the
preparatory embryological works of Rathke and others, the absurdity of
the Oken-Goethe theory of the cranium's being composed of vertebra;, while
at the same time he admitted its original metameric structure. This proved
a severe blow to Owen's archetype theory; from that moment the aged, cap-
tious anatomist became Huxley's enemy, all the more so as the latter had
already rejected the idealistic morphology. When, later on, old Owen de-
clared that the human brain has certain parts which no other animal can be
shown to possess and sought on these grounds to claim for the human race
a special position as towards the rest of the animals, Huxley proved in a
sharply critical way that the anatomical details of Owen's account were en-
tirely inaccurate, and this nullified all the latter's efforts to isolate man from
the animals.
Huxley embraces Darwinism
Nevertheless, in his youth Huxley was an upholder of the immutability
of the species, and an opponent of Lamarck's theories of evolution. Upon
490 THE HISTORY OF BIOLOGY
the advent of Darwin, however, he was one of the first to be convinced and
from that time onwards became one of the most zealous champions of Dar-
winism — its general agent, as he himself jokingly remarked. He took part
in the earliest controversy on The Origin of Species, contributing a paper that
proves how understanding and at the same time how independent was his
attitude towards Darwin's theory from the very outset. In the first place,
he has not Darwin's blind faith in the absolute dominance of the small
variations in nature. He cites as an example of sudden changes the oft-
quoted ancon or otter sheep of America, whose sudden appearance is a
well-known fact, and further, borrowing from Reaumur, the story of a fam-
ily that had a child with an excessive number of fingers and toes, which
phenomenon was afterwards inherited by its descendants. The arising of the
otter sheep is an obvious mutation; from the appearance of the supernumer-
ary fingers, again, conclusions might have been drawn in the spirit of Mendel.
This, however, was not done; even in the moderate form that Huxley gave to
his divergences from the true selection-theory, they attracted no attention;
small indeed would the variations have to be for the struggle for existence
and selection to have any material effect on them, and what interest could be
awakened by the story of the inheritance of six fingers? On the basis of such
exact observations of detail one came no closer to the theory of creation,
which, indeed, was the main idea at that time. Regret has often been ex-
pressed that Mendel's observations were published in such an out-of-the-
way place that no one noticed them; it is more than likely that the result
would have been the same wherever they had appeared; the fact is, the
time was not yet ripe for them. — However, Huxley's objections to the
master's theories were not numerous, nor were they bitter; he took far greater
pains to defend what good he found in them, which, indeed, was a very
great deal. And in contrast to Darwin himself, Huxley was a born controver-
sialist, with an ever-wakeful pugnacity, a never-failing promptness in reply,
an extensive knowledge of books, and a rare gift of putting the most in-
volved questions in a fluent and easily understood style. For the rest, his
polemics are always courteous; sceptic as he is, he confronts his opponent
with a supercilious, but not always a friendly, smile, and he never allows his
composure to be ruffled, nor himself to be reduced to silence. He particularly
enjoyed crossing swords with men of the Church, and on that battlefield
there were certainly to be found opponents en masse so long as he lived.
Among them was Gladstone, the great Liberal statesman, who was also
an extremely learned and highly conservative theologian. At one period
during the eighties Huxley entered into a controversy with him — the fore-
most biologist against the leading statesman in England at that time —
concerning the gospel story of the Gadarene swine, which were drowned
after the Devil had entered into them. Rather more urgent problems were
MODERN BIOLOGY 491
dealt with in his dispute with another famous politician, the Duke of Argyle,
who in a work entitled The Reign of Laiv had opposed Darwinism's lack of con-
formity to law in the sense given to it by idealistic natural philosophy. But
Huxley sought to influence the contemporary world of ideas also in a posi-
tive way; he enunciated the same social ethics that Darwin had taught and
that their age so largely embraced; he laboured to make the results of modern
natural science the basis of school education instead of the traditional classi-
cal languages, and he endeavoured by means of popular writings and lectures
to bring them to the knowledge of the general public. As a popular scien-
tific writer he is unrivalled for the clearness, warmth, and honesty of his style;
he never expresses a view that he cannot defend and never tries to disguise
the fact that the capacity of science for explaining phenomena is limited.
The same honesty he displayed also as a specialist. He once described a gelati-
nous substance taken from the bottom of the sea, which he thought was
a kind of undifferentiated, but living plasm, and which in honour of Haeckel
he named Bathybius haeckelii; when it was later discovered that the substance
was an inanimate precipitation, he frankly and boldly acknowledged his
mistake, which Haeckel found it rather difficult to admit. But Huxley was
also interested in purely philosophical problems; he was a great admirer of
David Hume, the famous sceptic of the eighteenth century, Kant's predeces-
sor as a critic of knowledge, and Huxley described his life and teaching in
a monograph. He himself represented his theoretical standpoint as agnosti-
cism, as a strict insistence upon the impossibility of knowing anything be-
yond the actual observations of the senses. And it must be admitted that he
succeeded in an unusually high degree in keeping free from the materialistic
dogmatism to which the opponents of the traditional ideals of thought and
religion are so easily addicted.
Gray on Darwinism
Among the professional botanists, also, Darwin at once found valuable sup-
■ porters. One of these was Asa Gray (1810-88), professor at Harvard Univer-
sity and well known as a leading writer on systematic botany and American
flora. Immediately after Darwin's appearance he came forward on his behalf
and in opposition to his own colleague Agassiz; the contributions he wrote
on behalf of the new theory in the course of a number of years he collected
in a special book, entitled Darwiniana. In his first review of The Origin of
Species he first of all attacks Agassiz's idea of species and then goes on to
point out that, even though Darwin was unable to prove his theory of de-
scent, he at any rate made an origin of species far less incredible than before.
Gray was at great pains to prove that Darwinism could be reconciled to the
belief in a personal God, and he likewise sets great store by the Darwinian
theory as being better adapted to finding an explanation of the finality in na-
ture than earlier theories; he expressly points out that Darwin rehabilitated
492. THE HISTORY OF BIOLOGY
the teleological explanation of nature — that is, the same as Kolliker
maintained and Huxley denied. To such divergent interpretations could
Darwin's theory give rise.
Darwin had also found a convinced supporter and a lifelong friend in
Joseph Dalton Hooker (1817-1911), a keen explorer in various exotic coun-
tries, a distinguished plant-systematist, and finally director of Kew Gardens
in London. Darwin was engaged in an almost constant interchange of ideas
with him during the many years he w^orked at his theory; when eventually
The Origin of Species was published. Hooker zealously defended the new the-
ory of descent, both in articles in the press and in his own more important
works. In particular, the introduction to his book on the flora of Tasmania,
which was published in the year after Tbe Origin of Species, represents a defence
of these doctrines, based partly on geographical arguments and partly on evi-
dence derived from classification, with special reference to the many phaner-
ogam genera which are so rich in species and, owing to their numerous
middle-forms, are so difficult to classify — Rubus, Rosa, Salix, and others.
Darwin's later researches into plant physiology were also supported to the
utmost by Hooker; in fact, of all the champions of Darwinism he was, on
the whole, in closest personal contact with its founder.
Development of Danvinism in England and Germany
It was from many different quarters, then, that Darwin won support for
his theory of descent; in fact, as has often been emphasized before, it be-
came the most widely cherished scientific idea of the age. In these circum-
stances it was obvious that many forces were destined to make for its further
development. Towards this end, however, it was possible to follow differ-
ent methods; of these there were really two that immediately came into use
— one in each of the two countries that came to be the principal centres of
Darwinism; in Germany recourse was had to the method which was only
hinted at by Darwin himself, of seeking fresh proofs of the descent theory
in comparative morphology and embryology, while in England his support-
ers followed the experimental and statistical method, in which Darwin him-
self had expressed the greatest confidence. These two courses will be described
in the following. In this connexion, however, we must first mention one
thinker who, although not a professional biologist, yet sought more con-
sistently than most people to apply the descent theory, in the form that
Darwin had given it, to all the phenomena of life, and who was regarded
by his contemporaries as the philosopher of evolution above all others.
Herbert Spencer was born at Derby in the midlands, in the year 1810,
the son of a schoolmaster. His parents were both Free Church people, but
belonged to different sects, and this lack of harmony induced feelings of
doubt in the son at an early age; in political radicalism, on the other hand,
he was fully in accord with his home throughout his life. He received a good
MODERN BIOLOGY 493
school education and especially distinguished himself in the exact sciences,
the classical languages having no attraction for him. He chose engineering
as his profession, distinguishing himself by a number of minor inventions.
His restless and insatiable desire for knowledge, how^ever, soon induced him
to abandon that career, and he resolved to devote himself to working out a
general scientific system. In order to carry out his purpose he studied many
different sciences, chiefly those of an exact character, and during that period
he earned a livelihood by writing for newspapers and journals. He never
received any public appointment and he consistently declined the honours
that were offered him, especially towards the close of his life, from many
quarters. In a constant struggle with poverty he lived in solitude, being
also during the latter part of his life a sufferer from a severe nervous afflic-
tion. He was ruthlessly radical, not only in his political views, but even
in his personal behaviour; he always gave his opinion straight out, and if
a conversation bored him, he put stoppers into his ears. In spite of his ill
health he lived to a good old age. When he died, in 1903, his body was
cremated without any funeral ceremony.
Spencer's idea of evolution
Herbert Spencer was not a specialist in biology, and his speculations on
biological problems have not advanced that science to any very great ex-
tent. He nevertheless deserves a place in the history of biology as a rare
example of a consummate and typical representative of that evolutional mode
of thought which was awakened to life by the general tendency of the times
in the middle of last century and which was promoted by Darwinism. He
is commonly called the most consistent philosopher of evolution which that
period produced — evolution forms the very groundwork of his system. In
its essential features this system was already pretty definite before the ad-
vent of Darwin; it was promulgated in a number of small articles in periodi-
cals, often characterized by masterly penetration and lucidity, afterwards
brought together to form an imposing work entitled A Systetn of Synthetic
Philosophy, which was the fruits of thirty years' work and which gives "a
broad, often too broad, development of what is recorded in the short trea-
tises" (HofFding). When Darwin produced his theory, Spencer associated
himself with it, although he interprets it after his own mind, and he became
one of the most influential promoters of the new doctrine of evolution.
Otherwise he is said not to have been in favour of extensive studies; he
preferred to think for himself and was very jealous of his independence.
Nevertheless, there is no doubt that Comte and his contemporary English
positivists exerted some influence upon him, and he himself admits that he
discussed biological problems with both Huxley and Hooker.
494 THE HISTORY OF BIOLOGY
Latv of differentiation
Of Spencer's shorter articles there is one dated 1851, "The Development
Hypothesis," in which he clearly and definitely dissociates himself from a
belief in the immutability of species; a hypothesis of creation is unscientific
because it is incomprehensible, and the probability is that the various forms
of life on the earth have been modified in the course of the ages by the in-
fluence of different external conditions of life. In a couple of other similarly
pro-Darwin essays, "Progress, its Law and Cause" and "Genesis of Sci-
ence," he gives a more general presentation of his evolutional theory, which
was afterwards further developed, in view of the selection theory, into his
great philosophical work. According to him, the function of philosophy is
to combine under one common standpoint the results achieved by all other
sciences: physics, chemistry, and biology, as also psychology and sociology.
This unity common to all sciences exists in evolution. All existence is evolu-
tion; the heavenly bodies are undergoing change, the earth was once incan-
descent and has since then gone through a series of evolutional forms, and
all things existing on it, both animate and inanimate, are doing the same;
the separate plant and animal individual is being evolved, just as species
and genera and humanity are being evolved, individual for individual and
generation after generation. The question of what "evolution" is, Spencer
has in such circumstances to try to get answered as exhaustively as possible.
In the above-mentioned treatise on the law of progress he endeavours to
formulate the answer from a biological standpoint; starting from the evolu-
tion theories of C. F. Wolff, Goethe, and von Baer, he finds in agreement
with them that the development of the individual proceeds from the homo-
geneous to the heterogeneous; out of the egg, which is uniform throughout,
both in structure and composition, is evolved an individual possessing vari-
ous parts and organs, which are the more differentiated the further the de-
velopment proceeds. This law Spencer believes holds good for everything;
the earth was once uniformly incandescent, but after having cooled off, it
acquired an increasingly different and varying surface; all living creatures
were originally primitive and homogeneous, but out of these primal forms
there has since been developed an ever greater multiplicity of life -forms; the
life of the human society offers the same picture, and differences in language
and other manifestations of intellectual life have similarly developed. But
whence is this differentiation produced? Spencer answers this question with
the contention that every cause invariably has more than one effect; if a
candle is lighted, it is one simple chemical process, but it produces a number
of different effects — heat, light, chemical products. Thus there are created
on the earth an ever-increasing number of phenomena. The whole of this
discussion on causality is, of course, a purely metaphysical problem; against
the theory of evolution on which it is based it may be remarked from a
MODERN BIOLOGY 495
biological point of view that Spencer deliberately threw himself into the
arms of the Wolffian epigenesis theory. If the standpoint of the preformation
theory is adopted, then the whole foundation of this doctrine of evolution is
destroyed. Now, in modern times, the egg is certainly not regarded as non-
differentiated; rather, with its numerous hereditary factors and the orienta-
tion given it from the very beginning, it is a tremendously complex structure.
Process of consolidation
At a later period Spencer tried also to expand his evolution theory. He sees
in it a process of consolidation; the egg-cell absorbs nutriment from sur-
rounding tissues, the embryo from the yolk of the Qgg, both under a proc-
ess of increasing consolidation. In the same way the celestial bodies have
been consolidated out of nebulous masses, and the human communities out
of scattered groups. Further, evolution may be regarded as a transition from
the indefinite to the definite, as indeed is demonstrated in the life of individ-
uals, species, and communities. But, above all, in his later years Spencer
began to realize that evolution does not always advance; it can also show
the exact opposite phenomenon, that progression and retrogression succeed
one another in evolution. This speculation suffers on the whole from the
attempt to bring all phenomena on the earth without exception under one
common definition, which in the circumstances becomes far too abstract:
it says too little because it is meant to embrace too much. The same fault
underlies the definition of life that is given in the biological section of Spen-
cer's system. Various characteristics of life are examined, and finally the
definitive characteristic is formulated thus: "Life is a continuous adjustment
of internal conditions to external conditions." The higher the life, the
stronger is the connexion between the internal and the external; the intel-
lectual life represents the highest degree of relationship between internal and
external changes. His detailed application of this theory of life offers little in
the way of interest; although controlled by Huxley and Hooker, it corre-
sponds but little to modern ideas. As an instance may be quoted the assertion
that life precedes organization in the matter in which it develops, whereas
in reality life and organization are indissolubly bound up in one another.
Limitation of the capacity for knotvledge
A LIKING for abstract conclusions has often been held to constitute Spen-
cer's chief weakness; it is in accord with the above-mentioned tendency to
bring together the most dissimilar phenomena in existence under one view-
point. He himself has defined knowledge as the bringing of every separate
phenomenon within the compass of a more general and previously known
one — the operation of muscle, for instance, is explained if one has a chance
of comparing it with the already known lever-mechanism — and he con-
tends that in consequence hereof the ultimate and most general phenomena
must remain incomprehensible because there is nothing more general with
496 THE HISTORY OF BIOLOGY
which to compare them. He repeatedly and with almost passionate emphasis
affirms that our capacity for knowledge is limited: what matter, force, space,
and time really are we shall never know, for our mind cannot grasp them;
we can only investigate the phenomena that our personal experience of them
educes. But for that reason Spencer also gives religion the right to hold its
own views on this " unknowable." Religious problems, however, have little
interest for him. He is all the more occupied with social questions, and it
is in this sphere that his evolution theory finds its most curious expression.
His belief in the progress of humanity is boundless and he is prepared to
apply to it unreservedly Darwin's theory of natural selection — that is, as
he himself says, that the fittest shall survive. The freedom of the individual
he places above all else: "Every man is free to do that which he wills, pro-
vided he infringes not the equal freedom of any other man." The State is
a survival from the primitive conditions of earlier ages, and its interference
with the life of the individual is purely wrong and merely hinders the opera-
tion of free selection. All measures adopted by the Government are worse
than if they were carried out by individuals; public poor-relief is expensive
and badly administered compared with private charity; State schools are
always inferior to private schools; in a word, the State should gradually be
done away with, but for the present it is necessary to maintain a police
force to ensure domestic security, and a military force to protect the country
from invasion, though on no account should there be compulsory military
service. So much the higher, then, must be the claims laid on private mo-
rality, and, in fact, Spencer claims much from it. He holds, in conformity
with his belief in the heredity of acquired qualities, that the intellectual
capacity of the individual becomes the common property of the race; the
quality of the intellect corresponds to certain structural conditions in the
brain; if the former is perfected, then the latter develop, are inherited by
the descendants, and thus benefit humanity. The aim of morality is to create
as much happiness as possible; happiness, however, must not be sought in
material prosperity — the more so as the latter leads to dishonesty. To be
allowed to contribute, in however small a way, towards the advancement
of general evolution should be the highest happiness to which the individ-
ual can attain. Morality thus benefits the community more than the individ-
ual, according to Spencer, as indeed according to the positivism of the age
as a whole. Both his and his contemporaries' limitation in this sphere lay
in an insufficient sense of the purely personal; he had but little sympathy
for the individual's longing for personal release from his confined and trying
environment or from his inner qualms of conscience; he thought that one
and all should take things calmly in the hope for better times to come —
which, indeed, seemed a far more likely prospect for the people of those days
than for those of our own.
MODERN BIOLOGY 497
As a matter of fact, Herbert Spencer himself lived to see the future of
the world darkened. The march of militarism, which he hated, went on
apace towards the close of the century; the colonization of tropical countries,
of which he also disapproved, was carried still further afield; while social-
ism, with its State production, must necessarily have been equally distaste-
ful to him. And even philosophy began in his lifetime to strike along paths
other than those he had marked out. But though his ideas are now for the
most part out of date, he will always be remembered as one of the most per-
sistent, disinterested, and courageous champions of the theory of evolution.
CHAPTER XIII
THE DOCTRINE OF DESCENT BASED ON MORPHOLOGICAL
GROUNDS. GEGENBAUR AND HIS SCHOOL
Leading position of Germany in biological research
IN HIS Geschichte der btologischen Theorten Radl declares that Darwinism was
born in England, but found a home in Germany. The statement is cer-
tainly justified in so far as, during the decades immediately succeeding
the first appearance of the descent theory, Germany came to take a leadmg
position in the sphere of biological research; here England and America
rapidly came under German influence, as also did Italy, while France which
kept itself isolated, nevertheless could not entirely avoid being influenced.
There were undoubtedly many reasons for this: on the one hand, the great
economic and technical development that resulted from the founding of the
German Empire, which in many ways proved beneficial to research, and,
on the other hand, the splendid manner in which the work at the German
universities was organized, which became a pattern for other countries, es-
pecially as a result of the careful and methodical guidance given by the
teachers to their pupils' theoretical studies, practical work, and scientific
production. And especially as far as biology in Germany is concerned, this
organization had reached a very high standard - chiefly in the sphere of
comparative anatomy - even before the appearance of Dar^vln. Originally,
of course, comparative anatomy had been based on idealistic morphology,
on the assumption that ideas formed the existing basis for the various forms
of life but we have already seen how this form of romantic natural philos-
ophy was gradually supplanted by a more realistic manner of viewing lite.
What Dai-winism gave to this realistic morphology was, as we know, a
hitherto lacking connexion in existence; common descent took the place ot
the common ideal types. The fact that it was the representatives of compara-
tive morphology in Germany who hailed the new doctrine with such deep
enthusiasm is explained by the insistent demand that they had of old f e t
for a uniform conception of nature, a heritage from the, at one time all-
prevailing, romantic philosophy. But it is just this never entirely eradicated
romantic element that gives to German Darwinism, with its application of
the descent theory to comparative anatomy, a character of its own. Again,
the general cultural situation in Germany at the time of the launching ot the
new doctrine must of course have had a considerable influence on the form
498
<;
n
Z
w
o
w
o
1-1
<
u
Pi
w
h-1
:0
O
>
H
P^
W
m
o
Q
P
MODERN BIOLOGY 499
it was to take. This is essentially bound up in the two names Gegenbaur and
Haeckel, each of whom in his own way represents a different side of the in-
fluence of Darwinism upon contemporary culture.
Carl Gegenbaur was born in 182.6 at Wiirzburg, of an ancient and well-
to-do family closely connected with the Civil Service. After being at school
in his native town he graduated at its university and applied himself, at
variance with his family traditions, to the study of medicine, with a view
to fulfilling his ambition to pursue scientific studies, in which he had early
shown a keen interest. With his natural bent for science, he at first derived
but little pleasure from his country's educational system; the gymnasium
in Wiirzburg was run by Jesuits and was conducted in the Jesuitical spirit;
nor were things much better at the University, until Kolliker arrived there
with Leydig as his assistant. From that time onwards biology received a
powerful stimulus, which was still further increased when, a couple of years
later, Virchow began his activities as a teacher there. Within a short time
these men made of Wiirzburg a nursery for biological research, and amongst
their pupils Gegenbaur at once took one of the foremost places. In 1851 he
wrote a dissertation for Kolliker and shortly afterwards accompanied his
master on his research expedition to the Mediterranean coast, a trip that
resulted in the young explorer's being ever afterwards attracted to compara-
tive anatomy. The immediate result of the voyage was a number of valuable
investigations into marine animals of various kinds, and the consequence of
these was a summons to a professorial chair at Jena in 1855. At this little
Protestant university, maintained by a liberal-minded Government, Gegen-
baur at once succeeded very well, although himself a Catholic; he had had
enough of the conditions prevailing at home, where the hospitals were un-
der ecclesiastical administration and the doctors were subject to clerical con-
trol. At Jena he gathered around him a host of like-minded friends and
pupils, chief among them being Haeckel. Here he worked out a scientific
system, which he afterwards applied throughout his life, and here too he
produced his finest works. In iSyx, however, he accepted an invitation to
Heidelberg, where larger resources were placed at his disposal and where
he afterwards laboured until the close of the century, when, owing to in-
creasing ill health, he resigned. He died in 1903, having been paralysed by
repeated strokes.
Gegenbaur was a forceful personality, a friend to his friends, and an
enemy to his enemies. As the founder of a school he is worthy of mention
with J. Miiller, but while the latter taught his pupils the method he had
invented and in theoretical questions allowed them to go their own way,
Gegenbaur permitted no divergence from the general principles he had once
and for all made his own. Moreover, he succeeded in inspiring his disciples
with such "boundless admiration" (Fiirbringer) that most of them were
500 THE HISTORY OF BIOLOGY
prepared for the rest of their lives to swear by the master's word. This
influence he won not through his lectures, for they were not very perfect
in form, but through the keen interest he showed in his pupils' work, pro-
vided that it followed the right direction; in the laboratory he was a friend
and comrade to his pupils and followed their careers in after life with never-
flagging interest. But he could never endure contradiction; as a controver-
sialist he was bitter and irreconcilable, although he invariably controlled
his language. At Jena he collaborated loyally with Haeckel, and the ex-
change of ideas that took place between them was mutual. After he removed
to Heidelberg, however, this co-operation ceased and even their friendship
cooled off, as Haeckel devoted himself to popular agitation, of which his
friend never approved. In his old age it was the Hegelian philosopher Kuno
Fischer who was in closest contact with Gegenbaur; he described the latter
as a deep thinker, which in its way characterizes him correctly.
Gegenbaur's first works came into being during his visit to the Medi-
terranean and comprise studies of the anatomy and evolution of various
marine animals; in particular, medusa; and other Coelenterata, Ascidia, and
worms were investigated by him during this period, many of them with
important results. Soon, however, he went over entirely to the study of the
Vertebrata. One of his earliest works in this field is an essay on the evolution
of the egg, published in 1861; in this he shows that all eggs of the Verte-
brata are simple cells; hitherto it had been supposed that the egg of the
bird, for instance, was a multicellular organ, whereas the granules in the
yolk were held to be independent cells. In this connexion he strongly sup-
ports Max Schultze's view that the cell need not necessarily possess a mem-
brane, but that the plasm and the nucleus are its principal com.ponents. This
investigation, which of all Gegenbaur's writings is perhaps of the great-
est value in the field of discovery, was followed by a long series of other
works, wherein he applies to different organic systems in the Vertebrata the
comparative method which he worked out in order to confirm Darwin's the-
ory, and the main principle of which is to discover by means of anatomical
comparisons the affinity between the animal forms due to descent. His finest
productions on this subject, taken as models for the whole generation of
research students, were his comparative studies of the skeleton, which were
brought together in the work Untersuchungen Tur vergleichenden Anatomk der
Wirbeltiere. Among these studies the most notable is the essay entitled " Car-
pus und Tarsus," in which he compares piece by piece the bones of the hand
and foot in different vertebrate animals, establishes their identity, and en-
deavours to reconstruct the form of extremity at one time possessed by the
ancestors of the Vertebrata. This primal extremity he terms in a subsequent
treatise " archipterygium " ; he holds that it has been developed out of the
gill apparatus and reconstructs the modifications by which have been evolved
MODERN BIOLOGY 501
therefrom the fins of the fishes on the one hand, and the motive organs of
the land-animals on the other. He supplemented this investigation with an-
other on the scapular apparatus and the pelvis, in which the bones of these
parts are similarly compared. The last and biggest section of this work is
called "Das Kopfskelett der Selaclner" and is described by Gegenbaur's dis-
ciples as the climax of his production. In it he examines and condemns after
the style of Huxley the old theory of the skull's being composed of verte-
bra:; instead he makes the cranium of the sharks the archetype from which
the same part in all higher vertebrates must have been derived; it comprises
in the sharks throughout their lives a cartilaginous capsule and is formed
of the same elements also in the higher vertebrates, while the latter's de-
finitive cranium is constructed with the co-operation of a number of covering
bones, originating in the skin. On the other hand, the visceral skeleton of
the head — gill-arches and jaw-bone — is compared with the ribs, so that
a part of the head, at any rate, possesses a segmented character. In conjunc-
tion with these skeletal investigations Gegenbaur also carried out a number
of comparative studies in the sphere of the anatomy of the nerves and mus-
culature and the organs of digestion. The entire results of his research work
he collected in that great work which was published towards the close of
his life, Verglekhende Anatomk der Wirbeltiere, in which he gives the most
complete expression to his ideas and aims. As parts of this work he added
his previously published text-books Grundxtige and Grundriss der vergleicben-
den Anatomie, which present his views, in concise form, and the method of
presentation and the contents of which have been imitated by many later
authors. In these, as also in his monographs, Gegenbaur's style is always
heavy and sometimes hard to understand, which his admirers held to in-
dicate depth of mind; nevertheless, consistency and set purpose are the most
conspicuous features in his scientific writings — which indeed explain their
success with his contemporaries.
Gegenbaur s general principles
In some essays written late in life Gegenbaur sets forth the principles
on which he considers that biological research should be carried out. To
him, comparative morphology is the essential science, not to say the only
road to the knowledge of life; and the final goal of this knowledge is the
determining of the mutual relationship of the different life-forms by dis-
covering their common origin. "The ultimate aim is phylogeny," he says
in an account of the relation of anatomy to ontogeny. After the fashion of
Darwin, he ascribes the actual formation of species to natural selection,
though, practically speaking, he does not discuss this problem, but con-
fines himself to tracing the individual organs back to common archetypes,
which he seeks in the lower organisms, in the vertebrates particularly in
the sharks. Investigations in homology with a phylogenetical purpose are
502. THE HISTORY OF BIOLOGY
thus the aim of his researches; the form and modification in the form of
the organs are all that interests him; physiological problems are thrust
aside, experimental investigations are unnecessary; histology is to him sim--
ply microscopical anatomy and he fails to understand its efforts to discover
the phenomena of metabolism in the elementary parts of the body. Even
embryology, which has nevertheless made such weighty contributions to
the theories of descent, is given no independent position, but is recommended
to adjust itself carefully to the results of the comparisons between the out-
grown organs. But, owing to the fact that these investigations into the prob-
lem of origin can, of course, never be verified, Gegenbaur's research work
proves in reality to be a theoretical speculation, which differs from that of
idealistic natural philosophy only in appearance, but not in reality. Gegen-
baur's archipterygium and Owen's archetype are practically alike fictitious,
only that the former is believed to have existed some time in the beginning
of the ages, whereas the latter had its existence located in the ideal world.
But Gegenbaur and his school are the last people to attribute unreality to
their primal types; provided one could once get the evolutionary series in
order and the gaps filled up with suitably reconstructed forms, it could be
urged that the primal type had as real an existence as if it had actually been
dug up out of one of the earliest fossiliferous deposits. Here is undoubtedly
demonstrated an intellectual contact with the romantic natural philosophy,
and Gegenbaur himself was without doubt influenced from that quarter; that
he as a thinker should have been approved by the surviving representatives
of Hegelianism was in this respect striking enough, and, as a matter of fact,
he himself has clearly expressed his sympathies for the romantic tendency
— Goethe's morphological schemes found in him a warm admirer; the for-
mer's and Oken's theory of the skull's being formed of vertebra is referred
to with unreserved acceptance, in spite of its not being tenable any longer.
The worst of such evolutional constructions, however, is that they are never
allowed to live long undisturbed, owing to the discovery of fresh facts, and
Gegenbaur's life's work has to a great extent had to suffer that fate. His
archipterygium theory was soon supplanted by another, which derived the
extremities from a lateral fin instead of from the gill-bones, and which perhaps
nowadays has most supporters. Besides, palasontological finds in recent years
have proved that the earliest amphibious types had seven digital bones in-
stead of the five that Gegenbaur assumed. Further, it has been shown in our
own day that the earliest fossil fishes possessed a cranium of bone, where-
fore the theory of the shark's cartilaginous cranium as a primal type is no
longer tenable.^ In certain other cases, too, he has been alternately right
^ For further reference to these questions see: Braus, "Die Enttvkklung der Form der Ex-
tranitdteri" in O. Hertwig's Handbuch der Engwkklungslehre der Wirbeltiere; Hans Steiner, "Die
Entwicklung des Vogelfiugelskelettes," in Acta Zoologica (Stockholm, 1912.); E. STENSio,Triassic Fishes
from Spitzbergen (Vienna, 192.1); and the literature referred to in those works.
MODERN BIOLOGY 503
and wrong, according as new facts have come to life. Thanks, however, to
his lirm convictions and will-power, Gegenbaur succeeded in compelling a
whole generation to follow his line of thought. Research on the subject of
origin was regarded as the most important function of science, and thus,
to quote his foremost and most independent disciple, Oscar Hertwig, hy-
pothesis was made the main point of evolution in science. And it must be
admitted that these theoretical speculations on the problem of descent have
had a highly stimulating effect upon morphological research; a number of
practical discoveries of the greatest value and of the highest significance for
the development of biology have been made by the Gegenbaur school. Even
to this very day comparative anatomy contains problems still unsolved and
still attracts investigators of worth. And though the purely speculative prob-
lems of descent do not, it is true, predominate to such an extent as formerly,
the presupposed common origin nevertheless still forms the basis on which
rest present-day homological investigations. But comparative anatomy
has certainly had to abandon its monopoly of biology and to recognize other
biological tendencies and methods as being equally justified in their
existence.
Furbringer on the system of birds
Of Gegenbaur's disciples the, majority naturally came from Germany, but
students also flocked to his institute from Scandinavia, England, and Russia.
Chief of these was Max Furbringer. Born in 1846, he eventually joined
Gegenbaur as a pupil at Jena-and accompanied him to Heidelberg as pro-
sector. Having for a time been professor at Amsterdam and Jena, he succeeded
his master at Heidelberg and faithfully preserved the latter's traditions. He
carried out comparative investigations in many fields; the excretal organs
of the vertebrates as compared with those of the Annelida, and the evolution
of the scapular regions are two of his best-known contributions to compara-
tive anatomy. He is chiefly to be remembered, however, for his Untersuch-
ungen ziif Morphologie und Systematik der Vogel, a monumental work in both
size and content. The first half of it consists of a comparative study on the
Gegenbaur model of the region of the breast, shoulder, and wing through-
out the whole order of birds. To this is added a general systematic section
setting forth the natural bird-system based on a comprehensive comparative
investigation of representatives of all the bird families. This system, which
is now universally accepted, has entirely re-formed the bird class; the old
orders are for the most part exploded — the owls, for instance, are trans-
ferred to the nightjars; the falcons and the vultures are placed next to the
petrels, herons, and storks — an example of how intensive anatomical in-
vestigations may give to the family relationships an entirely different value
from that of ancient tradition. One of Fiirbringer's advantages is that he
avoids the fanciful elements in the descent theories in which his age other-
wise abounds; he investigates the extant birds, but produces no reconstructed
504 THE HISTORY OF BIOLOGY
middle-forms and archetypes, letting the scientific material that actually
exists speak for itself. His method thus acquires a soundness that has given
to his results a lasting value. He died in 1910.
Hubrecht on phytogeny
Gegenbaur had a far more imaginative disciple in the person of the Dutch-
man A. A. W. Hubrecht (1853-1915), a professor at Utrecht. In his younger
days he was occupied mostly with the invertebrates, especially the worms,
but he afterwards devoted himself entirely to the evolution of the Mam-
malia. In this field he made valuable contributions by collecting material
in the course of expeditions in tropical countries and by investigating, with
special reference to their embryonic development, a large number of rare and
little-known animal forms. On the basis of this material he speculates deeply
upon the origin of the Vertebrata from lower animal forms, producing a
number of theories , that diverge considerably from what the earlier evolu-
tionists regarded as indisputable truth. The sharks, for instance, he places
for palasontological reasons in an isolated position in the system — that is
to say, in direct opposition to Gegenbaur's view — one of many proofs of
how fresh facts in this sphere have produced fresh difficulties, which it has
not been possible to solve with uniform results.
A complete account of the works produced by the pupils of Gegenbaur
or in his spirit would fill a volume; even in modern times comparative anat-
omy is largely under the influence of his method. Pupils had flocked to his
laboratory from all parts of the world. In the following chapter will also
be mentioned some pupils of Gegenbaur, who in their younger days followed
in his footsteps, but in later years struck out new paths of their own, some-
times with brilliant success.
Even the man who was in closest touch with Gegenbaur in his best
days and who exercised most influence on him, just as he was most in-
fluenced by him, went his ov/n way towards the end; this was Haeckel, a
man who gave to Darwinism a peculiar stamp, extremely characteristic of
the age, and who contributed much to its success, though perhaps still more
to bringing it into discredit.
CHAPTER XIV
HAECKEL AND MONISM
ERNST Heinrich Haeckel was born at Potsdam in 1834. His father had
taken part as a volunteer in the Prussian War of Independence against
Napoleon, and afterwards, having adopted a public career, he ad-
vanced to the rank of " Regierungsrat." His mother was the daughter of a
Civil Servant, who had been dismissed from his post and arrested for having
opposed the French conqueror. His parents' home, in spite of its bureau-
cratic character, had nevertheless preserved the liberal-minded traditions of
earlier times and the literary interests acquired in the days of greatness of
the German world of letters. Young Ernst received his school education in
a provincial gymnasium, where, as he himself says, mathematics were neg-
lected for philosophy and the classical languages. Even when grown up,
he still enjoyed reading Homer in the original and throughout his life de-
lighted in interlarding his writings with Greek terms. His greatest pleasure,
however, he found in nature, both in reality and in poetry; he was a keen
botanist and at the same time read Goethe's works, Humboldt's travels,
and Schleiden's popular writings. He was especially interested in Schlei-
den and was very anxious to go to Jena in order to be trained as a botanist
under him. After he had matriculated, in 1851, however, his father insisted
upon his going to Wiirzburg to study medicine. He spent two years there
and, in spite of his dislike for professional studies, devoted himself with
interest to anatomy and histology under KoUiker and to pathology under
Virchow. He then spent one year in Berlin studying under J. Miiller, whom
he regarded as his true master and who inspired him with a love for marine
research, particularly in regard to the lower animals. Having been for some
time assistant to Virchow, he went on an expedition to the Mediterranean
at the suggestion of Kolliker and Gegenbaur and collected at Messina ma-
terial for his first important work. Die Radiolarien, which resulted in his
being called, upon the recommendation of Gegenbaur, to the chair of zool-
ogy at Jena in i86i. There he worked until his resignation, in 1909, after
which he lived for another ten years, continuing his literary work until
he died, in 1919.
The life which falls within these dates is without doubt one of the most
remarkable during the epoch just closed; there are not many personalities
who have so powerfully influenced the development of human culture —
505
5o6 THE HISTORY OF BIOLOGY
and that, too, in many different spheres — as Haeckel. He has been much
disputed — now praised to the skies, now vilely abused. As a matter of fact,
it is not at all easy to grasp the true value of his life's work. No important
scientific discovery attaches to his name, and the ideas he promulgated are
largely borrowed from others. The works that once brought him fame are
now hopelessly out of date, but it must be admitted that much in them has
now been incorporated in our general knowledge. The idea of evolution, in
the form given to it by Darwin, found in Haeckel its most devoted cham-
pion; his personality and his trend of thought have set their mark on the
elaboration of this theory, especially on the continent of Europe, and they
are therefore worthy of closer examination.
Light is thrown on Haeckel 's early development by two collections of
letters, which have since been published, the one addressed to his parents
in his student days at Wiirzburg, the other to his betrothed during his Ital-
ian journey. This development is highly characteristic of the generation to
which he belonged and therefore explains in some degree how it was that
he acquired such an influence over his age. Young Haeckel at Wiirzburg
is by no means a German "corps" student of the ordinary type; on the
contrary, he was a very nice youth, abhorring duels and drinking-bouts,
diligently attending lectures and exercises, writing tender and affectionate
letters to his parents, regularly attending church, and comforting his lonely
hours with pious thoughts. True, he could cause his parents anxiety on ac-
count of his dislike for medicine and his propensity for unpractical dreaming,
but, on the other hand, he was always ready, with a somewhat rhetorical
and precocious eloquence, to confess his weaknesses to his old parents and
to promise to make them happy in the future. The most striking feature of
these letters is their Christian piety, which contrasts strongly with the
hatred that Haeckel felt for Christianity in later years; the youth expresses
his indignation against Karl Vogt and other "materialists" of the time in
terms that were afterwards used almost word for word against himself. It
is, of course, the opinions held in his parents' home that here recur — the
old-fashioned, serious, moral-religious atmosphere pervading the home of
a Prussian Civil Servant, with its literary and patriotic traditions. At Wiirz-
burg young Haeckel was enraged at the Catholic propaganda, which was
carried on at that time, during the period of reaction after 1848, with extreme
ruthlessness, and at the same time as his father was deploring the unhappy
political situation. In the letters from Italy the whole aspect is altered; that
was in 1859, ^^^ Y^^^ ^^ ^^^ liberation of Italy. Haeckel is full of enthusiasm
over Germany's unification and raves against her opponents, vassal princes
and Prussian junkers, who were serving the reactionary politics of Austria.
His religious attitude is now something quite different; Christianity has been
superseded by a worship of humanity in general, combined with enthusiasm
MODERN BIOLOGY 507
for the enlightened minds of classical antiquity and hatred against the
ecclesiastical reaction — a very common trend of thought at that time —
which found expression in many quarters in literature, as, for instance, in
the works of Haeckel's contemporary fellow-countryman Paul Heyse. His
biographers declare that Haeckel's religious change of front took place in
the course of spiritual struggles, but there is little trace of them in his letters;
it would appear more likely that with him, as with countless others, reli-
gious free-thinking was induced by political independence of thought; it
w^as difficult in those days to reconcile Christian belief and political liberal-
mindedness, owing to the Church's intimate connexion with the reactionary
forces in society and her obstinate resistance to all movements of reform.
Through his free-thinking, however, Haeckel lost that conviction which
had kept him going before, and he felt himself beginning to doubt the possi-
bility of penetrating any deeper into the essence of natural phenomena.
Haeckel embraces Danvinism
That guiding line for his thoughts which he thus lost Haeckel rediscovered
when he made the acquaintance of Darwin's theory. In Germany as in Eng-
land this theory had been received with mixed feelings; instances of this
have been given above. Haeckel at once became an ardent supporter of the
new doctrine; in it he found not only the means to understand existence,
but also the confirmation of the progress he desired to find in it. It was
mainly through his promulgation of it that Darwinism became a watch-
word for all supporters of the idea of a liberal-minded development in the
sphere of social and cultural life, and obviously an abomination to its op-
ponents, the clerical and conservative elements in the community. The course
of social development in Germany took an unexpected turn, however; the
unification of the country, the long-cherished dream of the free-minded, was
brought to reality through Bismarck, but in such a manner that the power
of the princes and the junkers was preserved. It was not thus that the lib-
erals had imagined things would turn out; their opinion now became divided;
the majority of them sided with the new work of unification and its leaders,
while a smaller group still insisted upon their demand for liberal-minded
social reforms. To this latter group belonged some of the leading scientists
in Germany, and among them Haeckel, although, living as he did in the
small town of Jena, he never took an active part in politics. It was with
all the greater enthusiasm, then, that he devoted himself to promoting this
radical development in the sphere of general culture, and he rapidly gained
a following of people with similar ideas to his own, who took up the strug-
gle against dogmatic conservatism in both the social and the religious sphere,
employing the evolution theory of Darwinism as their principal weapon. Of
course the authorities viewed with anything but friendly eyes this natural-
scientific opposition, with its social tinge; the employment of these hostile
5o8 THE HISTORY OF BIOLOGY
elements in the government service was opposed with all their power, and
many of the radical party lived for the rest of their days as independent
writers. Haeckel himself, however, was protected by the liberal-minded
Weimar Government from all unpleasant consequences and was thus en-
abled, though a professor, to take a leading part in the struggle. It goes
without saying that the natural-scientific contents of this doctrine were in-
fluenced by the political and social views of the antagonists, but, on the
other hand, this circumstance contributed towards making Darwinism popu-
ular and creating a widespread interest in its problems and arguments. Before
proceeding to describe the part Haeckel played in this struggle, however,
we must take a glance at the subject of research that he made his own and
determine how far his general scientific conclusions were based thereon.
His work on Kadiolaria
Haeckel began as a microscopist; when he was at Wiirzburg his father gave
him a microscope and he could not find words to express his delight at all
that he saw in it. In fact, both the papers he wrote for his degree were on
microscopical subjects — his dissertation on the tissues of the river cray-
fish, and an essay on the pathological changes of the venous system; two
school essays, the former worked out under the guidance of J. Miiller, the
latter under that of Virchow and noteworthy as being Haeckel's only spe-
cialized investigation in the sphere of the vertebrates — both papers credit-
able in their form and contents, but not very original. His appearance as an
independent investigator is marked by his monumental work on the Radio-
laria, which is without doubt his best. It is dedicated to the memory of
J. Miiller and is written in his spirit; he was, in fact, his foremost predecessor
in that field. It contains about one hundred and fifty new and carefully de-
scribed and illustrated species, as well as abundant material derived from
observations of their structure and mode of life. It makes what were at the
time valuable contributions to the problem of the biology of single-celled
animals, and moreover, the identity established by Max Schultze of the
protoplasm in the higher animals and the sarcode, which had already been
described by Dujardin, is hereby confirmed with fresh proofs. In connexion
therewith several cytological observations are quoted that are of consider-
able general interest — on the phenomena of currents and the manifestations
of assimilation in pseudopods and protoplasm, and also on the power of
cells to absorb solid bodies. Haeckel has observed how the blood corpuscles
in a mollusc absorb indigo-particles injected into the blood, but he did not
follow up this important fact any further, it being left to MetschnikofF a
couple of decades later to take up the subject and make it the basis of his
theory of phagocytes. In regard to classification, Haeckel tries to found
a natural system based on affinity; it is in connexion with this that he an-
nounces for the first time, though tentatively, his association with Darwin's
MODERN BIOLOGY 509
theory and endeavours to find a primal form from which the rest of the
Radiolaria could be derived. On the whole, however, the system was set
up in the traditional way.
Later on Haeckel continued the work on the Radiolaria, adding two
new parts (1887-8), and in the special section of these he describes several
hundred new species, which are splendidly illustrated. The complaint has
been made against his descriptions that they keep too much to the skeleton,
and against the illustrations that they are over-simplified, but on the whole
both text and illustrations compare creditably with the former volumes of
the work. The general section of the new work, on the other hand, is strongly
characterized by the natural-philosophical speculations which Haeckel had
produced in the mean while, and to which we shall revert later on.
His work on sponges
Another field that Haeckel made the subject of systematic research was the
sponges, of which he dealt especially with the Calcarea in his monograph
Die Kalkschwdmme, of 1871. In this work he has recorded his most consist-
ent attempt to create a Darwinistic classificational system — a true "nat-
ural" system based on descent, instead of the old "artificial" system. The
group had been very little investigated and the facts contributed by Haeckel
are of some importance, considering the age when they were published, al-
though they have, of course, undergone a good deal of modification as a
result of subsequent research. On the other hand, this natural system has
its curious features. The order of the Calcarea is divided into families ac-
cording to the shape of the canals in the walls of the sponge, and this mode
of classification has been retained by subsequent naturalists. The division
into genera, on the other hand, is based on the calcareous spicules of the
skeleton. These two features, the canals and the calcareous skeleton, are, ac-
cording to Haeckel, the only systematically employable elements, for their
form is inherited, whereas the artificial system has taken account of mouth
formation and colony or solitary life, which are dependent upon "adapta-
tion." No evidence, however, is offered in proof of these statements, and
it certainly does seem decidedly artificial to base the division into genera
upon one single character, without the slightest attempt to test the theory
by means of comparative morphology. Nevertheless, the system has its pe-
culiar interest as an attempt to separate entirely from the Linnasan system.
The terms hitherto employed have been entirely abolished; instead of " gen-
era" he uses "generic varieties," besides which there are differentiated and
nominated "specific, connective, and transitory varieties" or "initial, bind-
ing, and transitional species." One must admit the logical consistency of
this attempt to get away from Linnasanism; the latter rests entirely upon
the immutability of the species, and if it is once denied, it is necessary really
to set up a new system with a different idea of species. Haeckel's attempt.
5IO THE HISTORY OF BIOLOGY
however, has proved unsuccessful and has failed to gain the acceptance of
more recent systematists; in his own later systematic works he himself uses
the old traditional terms of genus and species, in spite of all the assurances
of Darwinism.
On the medusa
A THIRD subject in which Haeckel worked as a systematist is the medusa;
here, too, he summarized his results in a monograph of huge dimensions,
entitled Das System der Medusen (1879), containing a large number of newly-
described forms and a system of classification that in part is of some value.
In particular, the two main groups that he classifies in it, the Craspedota
and the Acraspeda, have been retained by later systematists. On the other
hand, it has been found upon examination that some of the diagnoses of
species are full of serious mistakes, which is explained by the fact that Haeckel
has in general a far keener eye for the demarcation of the large groups in the
system than for genera and species; careful detailed examination was never
his strong point.
There is still another group of life -forms which engaged Haeckel 's in-
terest and which perhaps appealed more to him than any other — namely,
the order Monera. To this order he refers single-celled organisms without
nucleus — that is, those formed of only a homogeneous mass. He has de-
scribed a great number of these — generally amoeboid organisms — many of
them with systematic validity. Nevertheless, the improved microscopy of
modern times has actually discovered in the majority of these a nuclear sub-
stance, either in the form of a single nucleus or divided into minute parti-
cles, and modern biology, which has learnt by experience to count the
nuclear substance among the essential components in a cell capable of life,
has in general presupposed the existence of the nucleus even in cells in which,
owing to its minimal dimensions or indistinct cell-content, it has not been
possible to confirm its existence. Haeckel, however, stubbornly held to his
non-nuclear Monera, the existence of which he regarded as an essential quali-
fication of that spontaneous generation by which he believed life to have
arisen, and which he looked upon as "a logical postulate for philosophical
natural science." This brings us to Haeckel's natural-philosophical specula-
tions — that part of his activities which, far more than his specialized re-
search-work, brought him both fame and ill fame.
The essentials of his opinion
As has already been mentioned, Haeckel declared his adherence to Darwinism
in his work on the Radiolaria. At a scientific congress in 1863 he expounded
Darwin's theory in a manner that considerably enhanced its success in Ger-
many. The lecture really comprised a brief summary of the Origin of Species —
of the doctrine of selection and the struggle for existence. In its essentials the
argumentation is Darwin's own, taken from the theory of domestic animals,
MODERN BIOLOGY 511
from animal geography and paleontology; but striking indeed are the radi-
cal conclusions that Haeckel draws in regard to the origin of man; they rep-
resent what eventually became one of his chief interests and immediately
caused a great sensation. There are two more peculiarities of Haeckelian
thought that come out clearly in this lecture: his political radicalism, which
induces him to call progress "a natural law which no human power, neither
the weapons of tyrants nor the curses of priests, can ever succeed in sup-
pressing" — the words were uttered just at the moment when the struggle
between Bismarck and his liberal opponents waxed hottest — and his pre-
dilection for the romantic natural philosophy, which makes him praise
Goethe, GeofTroy Saint-Hilaire, and Oken as "deep-thinking men possess-
ing prophetic inspiration," and as supporters of "philosophical theories of
evolution" foreshadowing Darwin. These elements — Darwin's theory of
evolution, political radicalism, and romantic natural philosophy — really
impress the whole of Haeckel's subsequent pronouncements with their char-
acter, whether they concern "general morphology," "cosmic riddles," or
"artificial forms in nature." The doctrine of natural selection forms the
groundwork, which he never takes steps to reconstruct or add to, however
great the progress made by research. His political radicalism mostly finds
expression in a violent hatred of priests and Christianity, but also, though
not so apparent, in opposition to the undue interference of government au-
thorities. The influence of romantic natural philosophy comes out most
clearly in his utter incapacity to grasp the relativity and limitations of
human knowledge, which Herbert Spencer among others so forcefully and
repeatedly emphasized; Haeckel's way of constantly trying to solve the
"riddles of the universe" is far more reminiscent of Schelling than of the
contemporary positivist trend of thought, just as his overbearing self-con-
fidence and his abusive polemics are more representative of romanticism than
of exact research. Thus through Haeckel's influence romantic natural philos-
ophy experienced a revival in the century of exact science.
His Generelle Morphologic
Haeckel struck his great blow for "philosophical scientific research" with
his Generelle Morphologie der Organismen, with its subtitle Kritiscbe Grundxiige
der mechanischen Wissenschaft von den entivickelten Formen der Organismen, be-
griindet durch die Desxendenztheorie , which was published in 1866. The first
part of the work was dedicated to Gegenbaur, the friend with whom he
had constantly exchanged ideas and who had inspired much of its contents.
The latter part of the book is dedicated to Darwin, Goethe, and Lamarck,
those "scientific thinkers who founded the theory of descent." As this trio
vras afterwards constantly referred to by Haeckel, it may be worth while
examining the combination more closely. Lamarck and Darwin may both
be regarded as founders of the theory of descent, although the latter, it is
5Ii THE HISTORY OF BIOLOGY
true, positively rejected the former's explanation of nature and was but little
concerned with its materialistic speculation. But the idea of seeing in Goethe
a precursor of a " mechanical science of the organisms ' ' certainly needs some
explanation. The great poet was universally looked upon by his age as an
idealistic natural philosopher; the biologists who acclaimed him did so under
that assumption and he himself had adduced " geistige Krdfte" as a cause of
the origin of and modifications in the life-forms, and otherwise also given
utterance to markedly spiritualistic views. Whence, then, Haeckel's asser-
tion to the contrary? The reason is no doubt to be found partly in Haeckel's
own natural-philosophical turn of mind, which could never be induced to
take the idea of "mechanism" in existence really seriously, and partly in
the position Goethe enjoyed in the cultural life of the period — his influ-
ence as a poet and a cultural personality, which was highly admired even
in Haeckel's home and circle, and which was opposed by no one beyond
the extreme-orthodox ecclesiastical authorities, who found free-thinking and
libertinism in his poetry, something which in its turn increased the sym-
pathy of the liberals for the really somewhat conservative poet-minister.
And the liberal opposition became once and for all one of the leading mo-
tives in Haeckel's system of thought.
The very choice of subject and the consequent title of the work — Gen-
eral Morphology — is also obviously borrowed direct from Goethe, who, in
fact, invented the word in question and from whom Haeckel also derived
the philosophical conception of morphology that he develops in the book.
For, strictly speaking, Haeckel was no professional morphologist in the
modern sense. He had till then worked almost exclusively on the classifi-
cation of single-celled animals; and in the comparative anatomy of the higher
animals, especially the Vertebrata, he practically never carried out any spe-
cial investigations, at least none of which the results have been published.
That he nevertheless based his theoretical speculation not on classification,
as Darwin himself did, but on morphology, was no doubt due, as hinted
above, to his admiration for Goethe, but also, of course, to the influence
exerted on him by his friend Gegenbaur, who was no doubt responsible for
the best contributions of facts in the work. But a speculatively inclined
student who concerns himself with second-hand knowledge will, of course,
easily succumb to the temptation to let his imagination get the better of
his critical sense — a fact that finds strong confirmation in Haeckel.
His ternary division of nature
The General Morphology begins with a chapter on the relation between mor-
phology and other sciences. First comes an assertion that every natural object
possesses three qualities: matter, form, and energy or function. In connexion
with this idea natural science is divided into three disciplines: chemistry,
or " Stojflehre," morphology, a.nd physics, or '^ Kraftlehre." Then the knowledge
MODERN BIOLOGY 513
of inorganic nature is divided into mineralogy, hydrology, and meteorology;
and biology is divided into xpology, -protistology, and botany. Thus we have
here four threefold groups, all extremely ill-grounded. One is tempted to
assume that it is really Schelling's romantic ternary mysticism haunting
him here — though, of course, indirectly and unconsciously. The division
into plants, animals, and protista is, of course, entirely useless, nor did it
ever succeed; instead of one vague line of demarcation such as that between
plants and animals, we here get two. Then the aims and means of morphol-
ogy are described; the aim is a mechanically causal explanation of the forms
and phenomena of life, whereby a "monistic" explanation of the universe
will be made possible, which indeed, it is declared, is already so in the
other natural sciences, but in biology is for the time being replaced by a
"vitalistic" and "dualistic" view, the incorrigibility of which is depicted
in vivid colours. The means of attaining this monistic explanation of nature
is declared to be by "philosophical thinking," by the aid of which, facts
should be capable of interpretation, whereas the mere observation of nat-
ural phenomena is deeply despised. As a matter of fact, this philosophizing
constitutes Haeckel's great weakness, which gradually induces him to aban-
don exact research. The insistence upon interpreting the phenomena of life
according to purely mechanical laws is in itself fully justified; physiology
had already pursued that method before Haeckel's time, and his claim that
the other branches of biology should follow its example was quite reason-
able. But Haeckel's great mistake lay in his refusal to realize and acknowl-
edge the limited possibilities of the mechanical explanation of nature. He
certainly admits in one passage (p. 105) that the human capacity for knowl-
edge has its limits: that we cannot reach the ultimate grounds for a single
phenomenon, and that the origin of a crystal down to its ultimate causes is
just as inexplicable as the origin of an organism. But he does not stop for a
moment to think that in such circumstances natural philosophy should en-
deavour to determine these limits and see that they are not exceeded. Shortly
after the above admission he confidently asserts that no essential difference
between animate and inanimate exists; after making a close comparison he
comes to the conclusion that the crystal and the living cell are in all respects
comparable, as to their physical and chemical composition, their growth and
individuality. The restriction that should follow from the limitation of the
human capacity for knowledge is entirely forgotten. The memory of it cer-
tainly reawakens now and then, but, generally speaking, he entertains a
blind faith in the power of "mechanical causality" to explain anything
whatever.
Mechanical interpretation of nature
What Haeckel chiefly bases his conviction upon as to the unlimited possi-
bilities of the mechanical explanation of nature is Darwin's theory. His
514 THE HISTORY OF BIOLOGY
enthusiasm for it knows absolutely no bounds; once he assures us outright
that, thanks to this theory, there is now not a single fact in organic life
that cannot be explained, although many are still unexplained. This en-
thusiasm, indeed, he shared with the whole of his generation; in a previous
chapter light has been thrown upon the hopes that the selection theory
aroused on its first appearance; the fact that in Haeckel they reached such
dizzy heights was due, of course, to his personal temperament, in which
enthusiasm, a naive self-satisfaction, and a blind confidence in the correct-
ness of his own ideas had been the predominant features since his youth.
Otherwise, he desired to a certain extent to modify the selection theory
itself, in so far as he would define more precisely the actual term "struggle
for existence." He urges the exclusion of all conditions belonging to sur-
rounding nature; the competition with other living creatures is all that
should be considered in this connexion. He further maintains the existence
of a competition ivithin the individual, between its various parts — that is,
an adaptation of the theory of correlation to Darwinism, which was later
developed in certain respects by others. And, finally, Haeckel insists, far
more emphatically than Darwin, upon the transformation of the individual
through the influence of environment and the inheritance of the modifica-
tions thus brought about; he defines evolution as a co-operation between an
" innerer Bildungstrieb" — heredity — and an " dusserer Bildungstrieb" — the
influence of environment. These expressions, which have a very natural-
philosophical and not a very mechanical sound, he borrowed, as he himself
admits, from Goethe's Pfianzenmefamorpbose, which he considers represents
Darwinism in mice, and which to his mind still forms the basis of plant
morphology — a view which at that time was shared by only a few sup-
porters of natural philosophy, but which has been repeated on Haeckel's
authority up to modern times by literary historians and other non-profes-
sionals. For the rest, he attributes to Darwinism an infinite mass of new
determinations, with their attendant terminology. Haeckel almost surpasses
Linnasus in his mania for classifying and naming, but he is entirely lack-
ing in the incomparable gift for form that the great systematist possessed;
most of his categories and nomenclature have not survived their originator,
although a number of them have been universally adopted, as for instance,
the terms "ontogeny" and "phylogeny," the former denoting the indi-
vidual's, the latter the race's development, and " cecology," as an expression
denoting the relation of living beings to their environment. Utterly absurd,
on the other hand is his " promorphological " classification of the life-forms
according to a symmetrical plan intended still further to confirm the alleged
similarity between the structure of crystals and organisms; the details of
this system, which, as a matter of fact, give evidence of a very superficial
knowledge of the foundations of crystallography, may be compared with
MODERN BIOLOGY 515
Oken's wildest flights of imagination; infusorians, pollen granules, corals,
flower-spikes, are cited, amongst other things, as examples of the various
supposed crystal-symmetrical forms. Undoubtedly more successful is the nat-
ural system that he afterwards sets up for the organisms, wherein is employed
for the first time the method, so frequently used since, of representing by
means of a graphic chart in the form of a genealogical tree the mutual agree-
ment of the different life-forms, as if derived from an assumed common origin.
Haeckel has certainly had to endure a good deal of chaff for his genealogical
trees and they will not, of course, bear too close examination, but it cannot
be denied that the method itself has proved of good service to scientific
works aiming at a natural system; we need only mention how Fiirbringer
employed it in his great work on the birds. Here, as in many other respects,
Haeckel has had a rousing and stimulating influence on subsequent research.
Haeckel idenfifies spirit and matter
The genealogical tree that now, as henceforth, interested Haeckel most is,
however, that of man; already at this stage he sets forth the ideas concern-
ing it that he was later to develop still further. From man he proceeds to
the universe and God, and now makes the entirely unexpected assertion that
"no matter can be conceived without spirit, and no spirit without matter."
It is hard to make out how this idea is to be reconciled with his earlier as-
surance that every natural phenomenon, both animate and inanimate, can
and is to be explained mechanically; ever since the days of Galileo, indeed,
all spirits have been outlawed from mechanics. Haeckel, nevertheless, makes
use of his spirit-matter to decree unity between God and nature — a unity
which denotes true monism and which admits of a true divine worship. It
is again from Goethe, of course, that these pantheistic reveries are borrowed,
so that in this first philosophical work of Haeckel's, romantic idealism has
the last word.
Generelle Morpbologie, which in Haeckel's own views is his principal
speculative work, had but little success; only one edition was published.
Darwin, it is true, was delighted, although he complained mildly of the
vehement style in which the book was written, but the German biologists
were enraged at the natural-philosophical daring, the dilettante treatment
of detail, and the scurrilous language. After some years of silence, however,
Haeckel resumed his natural-philosophical activities, this time in a more
popular form, with the result that he was extremely successful with both
his series of lectures Naturliche Schopjungsgeschichte (1868) and his Anthropo-
genie oder Entivicklungsgeschichte des Menscben (1874); ^^^ former work espe-
cially became extraordinarily popular, being translated into many languages,
and it really represents perhaps the chief source of the world's knowledge
of Darwinism. It reproduces the ideas and the arguments from Generelle Aior-
phologie, but in an easier style and excluding his extensive speculations on
5l6 THE HISTORY OF BIOLOGY
symmetry. He gives instead a special account of the descent of man, which
Haeckel regarded all along as the centre point of the theory of evolution
and of all science in general. This problem likewise represents, as the title
indicates, the subject of his Ant hrofo genie, a work of far greater significance
than the History of the Creation and certainly the one in which Haeckel has
set forth h'is most brilliant and most important ideas — those of his that
most deeply affected the development of biology. The intention of the work
is to give a comprehensive idea of the origin of man, based on the evidence
of morphology, embryology, and palaeontology. Haeckel definitely takes as
his starting-point his well-known "biogenetical principle": that the on-
togeny not only of man, but also of every living creature is a recapitulation
of its phylogeny; "the development of the embryo is an abstract of the his-
tory of the genus." This idea in itself is not new; as we have seen, it had
already been propounded by Meckel, and Darwin gave it an important place,
although it was formulated in summary fashion, in his Origin of Species. It
was then taken up and further elaborated by Fritz Muller (1811-97), one
of the more peculiar representatives of biology during last century.
Fritz Muller on the development of crayfish
Born in Germany, Fritz Muller had studied medicine — among other things,
biology under J. Muller — but afterwards went out to Brazil, where he re-
mained for the rest of his life in various occupations and with varying for-
tunes. Having from the very beginning been entirely won over to Darwin's
theory, he resolved to prove it by applying it in detail to a suitable animal
group, for which purpose he chose Crustacea, which in his adopted country
exist in a multitude of forms. He paid special attention to the different types
of development to be found in closer related forms within this class: the
river crayfish creeps out of the e§,g like its parents; the crabs have one or
two larval forms, while the prawn has many — a nauplius stage similar to
the larvas of the lowest Crustacea described under that name, a 7j)ea stage,
like that of the crabs, and a my sis stage, like the perfected form of the schiz-
opod crayfish. Various other Crustacea likewise possess peculiar metamor-
phoses, especially the strangely formed parasite crayfish, whose early stages
resemble those of the independently living Crustacea. All these facts, espe-
cially the fact that the larvas of certain higher Crustacea resemble the fully
grown individuals of lower Crustacea, convinced Fritz Muller that the evo-
lution of the individual is a "historical document," which is sometimes
effaced, owing to the development's striking into a more and more direct
path from the tgg to the fully grown creature, and which is sometimes
"counterfeited, owing to the struggle for existence that the independently
existing larva; have to maintain." A case such as that of the prawn he re-
gards as typical; the prawn's ancestors in past ages possessed the form that
its larva; now possess, and that, too, in the same sequence as that in which
MODERN BIOLOGY 517
the larva: now succeed one another, whereas the history of the river crayfish
has been obliterated and that of a number of other Crustacea has received
fresh contributions in respect of form.
This theory, which Fritz Miiller expounded in 1S64 in a paper entitled
Fur Darwin, aroused Haeckel's ardent enthusiasm. To him it became a "prin-
ciple for the origin of life," the main support of the theory of descent and
a particularly weighty argument in the controversy over the struggle for
man's "natural creation." It was then chiefly to human evolution that he
sought to apply the theory and in his Anthrofo genie, as also previously in
his Naturliche Schopfungs geschichte , he works out the embryonic development
of man from the es,g to birth with a view to collecting proofs of the condi-
tions governing man's descent and affinity. Haeckel was never a specialist
in embryology and its points of detail were of no interest to him in them-
selves, but only in so far as they could serve as evidence to prove the descent
of man. His ideas of embryology could in such circumstances only be one-
sided and deficient; the professional embryologists offered serious objections
to them, which he either affected to overlook or else answered with per-
sonal abuse. Complaints were made especially against his illustrations,
which, contrary to usual practice, he hardly ever borrowed from mono-
graphs on the subject, but drew himself. Being designed exclusively to prove
one single assertion, his illustrations were naturally extremely schematic and
without a trace of scientific value, sometimes indeed so far divergent from
the actual facts as to cause him to be accused of deliberate falsification — •
an accusation that a knowledge of his character would have at once refuted.'
Haeckel's theory of germinal layers and gastrcea
Two specially remarkable details in Haeckel's doctrine of the biogenetical
principle are the theory of the germinal layers and the gastrasa theory. We
have previously described the investigations into the embryonic germinal
layers carried out by Pander, von Baer, Remak, and others, and also how
Huxley compared dermal and intestinal layers in the medusa: with the ger-
minal layers in higher animals. Besides these facts Haeckel had for material
on which to work his own researches into the Calcarea, the embryonal de-
velopment of which he had studied. On all this he now bases the theory of
the origin of the animals, and especially that of man; since man originates
from a single cell, the ^gg, then in the beginning of time the original form
out of which the human race has evolved must also have been a unicellular
animal. Out of the egg-cell there is developed by segmentation a cell-group;
^ It is nevertheless difficult to understand such an action as this : allowing in his Naturliche
Schopfungsgeschichte (ed. i, p. 241) the same cliche, reproduced three times, to represent an egg of
a man, an ape, and a dog. This absurdity was removed from subsequent editions, albeit only
after Haeckel had rewarded with abuse those who pointed out the fact; and the incident was for
ever afterwards a theme on which his enemies constantly harped.
5l8 THE HISTORY OF BIOLOGY
this stage the primal forms of the higher animals and man have also passed
through, and during that period they have resembled such cell-colonies, as,
for instance, the Vol vox. Out of the simple cell-mass there evolves in the
sponges, by means of invagination, a stage of development v^ith double
walls, a gastrula, which corresponds to the simplest form of an animal pos-
sessing an intestinal canal; the original form of the higher animals must
likewise have passed through this stage. This original form common to all
higher animals is called gastrasa. From each of the walls of the gastrula
there splits off through segmentation a fresh layer; these two secondary
layers combine and form the mesoderm, which gives rise to the muscula-
ture and various other organs in the higher animals. This process has also
taken place at some time or other in the primal form of the higher animals,
and therefore all these three layers and their derivatives are homologous
throughout the entire animal kingdom.
Importance of Haeckel's bio genet ical principle
This evolutional theory is undeniably Haeckel's most brilliant and most
important contribution to the history of biology. O. Hertwig was right in
saying that for fifty years biological literature was under the influence of
this idea; the abundant facts that were amassed on the subject of embryology
during this period were mostly intended to confirm the biogenetical principle
or the "recapitulation" theory, as it has also been called, and biologist;,
strained every effort to apply it to every detail in the development of th;
embryo. And the application was "strained" in the fullest sense of the word.
Haeckel knew from the outset that the gastrula stage of the mammals is not
formed through invagination, as the theory claimed, but through delamina-
tion, or splitting off; he consoled himself, however, with the thought that
in the lancet-fish invagination generally takes place, and from this primi-
tive animal he derives the Mammalia, with the assertion that their gastrula
form is due to later adaptation — to the "falsification" of documents, of
which Fritz Miiller had spoken. He also explains a number of other facts
of a similar kind according to the same method. Matters became still worse
w^hen the embryologist His came forward with an attempt to explain the en-
tire cause of embryonic development on purely mechanical grounds. Haeckel
was furious and replied with a shower of abuse, quite forgetting all his
own utterances, in which he insisted upon a mechanical explanation of na-
ture. In reality this mechanical, or, in other words, physiological, side of
embryonic development is of very great importance, though Haeckel quite
overlooked the fact in his anxiety to explain natural creation; later on, how-
ever, it received all the greater attention. But, even apart from this, time has
dealt hardly with Haeckel's ontogenetical theories. The gastrula forma-
tion by means of invagination has proved far less general than Haeckel
believed — inter alia, it is lacking in most of the Coelenterata — and the
MODERN BIOLOGY 519
far-fetched homologization of the germinal layers has been considerably re-
stricted, the same organs in a number of different animal forms having been
found to possess an entirely different origin. In particular, the mesodermal
formation has now been resolved into a number of different processes. In
fact, the entire " biogenetical principle" is nowadays severely challenged,
even as a hypothesis; in the vegetable kingdom it has received no confirma-
tion, which is indeed strange for a theory proposed to hold good as a gen-
eral explanation of life, but even those zoologists who in general give any
support at all to the recapitulation theory do so with considerable reserva-
tions, called for by the results of modern hereditary research and experi-
mental biology. Nowadays one does not compare without question, as they
did in Haeckel's time, the ideas of similarity and affinity ; similarity can
demonstrably arise through the influence of very different factors, and it is
preferred to follow His in seeking for the mechanical conditions governing
the development of form instead of seeing therein resemblances to the ani-
mal life of past ages. But this should not involve our depreciating Haeckel's
influence on the development of embryology; it was his theory which evoked
that interest in those phenomena that brought about the immense revival
of this form of research, lasting up to the present day. In this connexion we
may remember von Baer's words that "inaccurate but definitely pronounced
general results have, through the corrections which they call for and the
keener observation of all the circumstances which they induce, almost in-
variably proved more profitable than cautious reserve." It is just herein that
Haeckel has benefited his science most; here he has made his most important
and historically most yaluable contribution. But with it he gave all that
he had to give; the years that he lived afterwards produced nothing to in-
crease his reputation, but detracted much from it.
His ivork on ' ' peri genes is
For as early as in his Antbropogeny Haeckel displays his increasing weakness
for vague and profitless speculations. Talk of a mechanical explanation of
nature is certainly kept up, but it becomes more and more empty w^ords,
while the spiritual qualities of matter appear increasingly in the foreground;
energy and soul are now consistently identified, and are generally denoted
by the term "energy," in a manner which testifies to his absolute contempt
for the simplest grounds of physics. And this fault is still more intensified
in a treatise published in 1875 entitled Die Ferigenesis der Plastidule, the nat-
ural-philosophical confusion pervading which it is truly difficult to repro-
duce in a summary. The title itself is supposed to mean " the wave-production
of life-particles" and this is intended to explain the same phenomena as
those upon which Darwin tries to throw light by means of his pangenesis
theory — that is, heredity and adaptation. The pangenesis theory fails to
satisfy Haeckel, who instead endeavours to explain heredity by an analysis
52.0 THE HISTORY OF BIOLOGY
of the molecules, or plastidules, as he calls them, of living matter. Life is
due to their atomic structure, and " jedes Atom besitxt eine inhdrente Summe von
Kraft und ist in diesem Sinne beseelt." Energy and soul are thus identified anew,
and this having been done, all difficulties disappear. Haeckel now explain^
reproduction, as always, with the old definition: growth over and above the
individual; heredity is a transmission of the motion of the plastidules, and
adaptation a change in this motion. This certainly sounds somewhat me-
chanical, but some pages further on we suddenly find a new definition:
"Die Erblichkeit ist das Geddchtnis der Plastidule, die Variabilitdt ist die
Fassungskraft der Plastidule." And yet still later we are told that "das Ge-
ddchtnis"' is a transmitted motion. It would, of course, be superfluous to judge
these fancies according to scientific standards ; Haeckel himself admits that
he got the idea of "the memory oi" the atoms" from Goethe's famous ro-
mance Die Wahlverivandtschaften, and indeed the whole plastidule theory
sounds like a romance; in producing it Haeckel had abandoned himself en-
tirely to romantic natural philosophy and there he remained for the rest of
his life. The phenomenon might seem to have only a psychological interest
and might be passed over with a reference to Haeckel's esthetic turn of
mind — he was, in fact, something of an artist, a gifted dilettante in water-
colour painting and an admirer of beauty both in art and in nature — but
it might also be pointed out that a pioneer in science may be considered
justified in entertaining some strange thoughts on general problems — this
has been acknowledged throughout the ages. Yet this does not explain how
it was that this speculative side of Haeckel's activities should have proved
capable of creating such an extraordinary sensation among his contempo-
raries — that people should have been so loud in their praises and in their
abuse. This point demands an explanation by itself, wherefore we must cast
a glance at the political and social conditions of the time.
Political radicalism of the Haeckelians
In Germany the seventies were a somewhat restless decade; the recent vic-
tories had certainly confirmed Bismarck in his power, but he nevertheless
had opponents in two directions: the Catholics, whose ultramontane politics
were regarded as a menace to the unity of the Empire, and the interna-
tional labour movement, which had recently found expression in the com-
munal riot in Paris, that had so scared the world, and not least Germany,
where some attempts against the lives of distinguished people were placed
to its account. In such circumstances the liberal-minded apprehended a fur-
ther reign of terror, and the friends of domestic peace still further social
upheavals. And Darwinism in particular, which indeed had from the begin-
ning been strikingly characterized as a theory of progress, through Haeckel's
boisterous attacks on the authorities of State and Church and through his
dogmatic description of the contrast between the doctrine of creation and
MODERN BIOLOGY 5x1
the theory of evolution, had been suspected by the conservatives and looked
upon as a socially dangerous hypothesis, the truth of which, moreover,
could be disputed on the grounds of the plastidule theory and similar ideas
contained in it. The exchange of ideas on Darwinism became in these cir-
cumstances more and more lively; round Haeckel there gathered a crowd
of young naturalists who preached the new doctrine with enthusiasm.
Among them may be named A. Brehm, the author of the universally known
work Animal Life, F. von Hellwald, known as a geographical writer,
G. Jager, famous for his curious hygienic theory, and others. Since the uni-
versities were mostly closed to them, they carried on their agitation by means
of popular lectures and polemical writings, in which they expounded their
views, willingly associating their natural-scientific radicalism with a po-
litical radicalism, and with this party the old radicals Vogt and Biichner
associated themselves. But the new theory claimed also politically con-
servative adherents, as, for instance, Du Bois-Reymond; he had, it is true,
embraced Darwinism with enthusiasm and had declared that its appearance
had freed biology from all explanations of the vexed problems of final-
ity, but at the same time he had expressed disapproval of "Haeckelism."
In his above-mentioned lecture on the limitations of our knowledge of
nature he had uttered a warning against belief in a possibility of definitely
solving the riddles of nature and life. Haeckel, who, it will be remembered,
had nevertheless himself admitted the limitation of man's capacity for
knowledge, became enraged at the word " ignorabimus,'' in which he scented
political reaction. The foreword to his Anthropogenie is directed against it
and treats the expression entirely politically. The situation became still more
tense some years later, when the Prussian Government was engaged in draft-
ing a new educational law, the provisions of w^hich were bound to affect the
future of science in Germany. Then Haeckel came forward at a scientific
meeting at Munich in 1877 with an address on the relation of the evolution
theory to science in general. In it he presented his old theories, including
the plastidule hypothesis, and expressed the assurance in connexion with
them that biology, as conceived evolutionally, is not an exact, but a his-
torical and philosophical, science, and as such aimed at uniting natural-
scientific research with the psychical sciences and thus forming the basis
for a uniform view of life, which would gradually reconstruct the whole
of human existence on general humanitarian lines, and which should there-
fore constitute the foundations of all education.
Virchow opposes Haeckel
This proposal was opposed by Virchow in a speech in which he points out
all that is hypothetical and unproved in Darwinism, and on these grounds
he uttered a warning against incorporating it in a scheme of school educa-
tion, for such a program should only concern itself with indisputable proofs.
52.1 THE HISTORY OF BIOLOGY
Virchow's speech was greeted with cheers by the conservatives; as a matter
of fact, its criticism of Haeckel's fantastic ideas was justified, but its peda-
gogical program was of doubtful value; it might reasonably be asked what
would be left if all hypothesis were banned in the schools — every explana-
tion of nature is fundamentally hypothetical, and much of the results of
historical research rests, of course, upon disputed facts. And even more un-
acceptable sounds his passing reference to the spiritual affinity of Darwinism
to socialism — a denunciation which at that time was equivalent to an ac-
cusation of high treason. Shortly after^vards the Prussian Minister of Edu-
cation sent round a circular strictly forbidding the schoolmasters in the
country to have anything to do with Darwinism, and in the new educational
law biology was entirely excluded from the curriculum for the highest
classes in the schools, with a view to protecting schoolchildren from the
dangers of the new doctrines. Haeckel replied to Virchow's speech in a pam-
phlet, Freie Wissenschaft und freie Lehre, in which he again formulates the
antithesis "Creation — Evolution," brings forward "certain proofs" of the
correctness of the theory of descent, declares that cell-psychology can be
traced to Virchow's own ideas, and finally urges the freedom of education
and Darwinism's independence of the political questions of the day. His reply
was hailed with enthusiasm by the free-thinkers and it is easy to realize
the eagerness with which the friends of the freedom of thought and word
must have gathered around him in spite of his many delusions, when such
measures as the school regulations mentioned above were adopted by the
opposite party. All the more so as the outcome proved Haeckel's justifica-
tion; Darwinism might be prohibited in the schools, but the idea of evolution
and its method penetrated everywhere, in historical research and linguistic
studies, and even in the scientific treatment of religious documents and reli-
gious history. And to this result Haeckel has undeniably contributed more
than most; everything of value in his utterances has become permanent,
while his blunders have been forgotten, as they deserve.
Victory of Darivinism
During the eighties the dispute as to the justification of Darwinism died
down; Haeckel himself spent most of this period in studying the Radiolaria,
and his partisans likewise began to pursue other activities. Instead, that
decade was to witness the undisputed domination of comparative morphol-
ogy in biological research and training; it was at a time when Gegenbaur's
and Haeckel's ideas universally prevailed without opposition and were ap-
plied to various groups of the animal kingdom. But the results were in no
wise what Haeckel had anticipated. Instead of simple and easily compre-
hended proofs of the indisputable validity of Darwinism, the younger gen-
eration of scientific students found masses of involved facts, which only
contributed to confuse the biogenetical principle, the gastrasa theory, and
MODERNBIOLOGY 5 13
the other "natural hiws." This was not at all what Haeckel had expected.
Self-confident by nature and spoiled by the successes of his earlier years, he
was lost amongst all these developments; intensive study of detail had never
been his strong point, and the minute methods and detailed observations
of the young morphologists aroused his keen opposition. In a letter dating
from this period he expresses the opinion that modern morphologists in
general, and " Querschnittler und Anilhifdrber'' in particular, possess far less
"logical schooling" than the systematists of the old school. And, as always,
special research was followed by a waning interest in theoretical specula-
tions; instead of paying attention to Haeckel's watchword — either crea-
tion or evolution — students preferred to leave the theories to their fate
and to go over to practice. When, then, even Haeckel's favourite idea of
man's origin from the higher apes and his affinity to the gorilla and the
chimpanzee began to be doubted by scientific students, who found man to
be in anatomical respects highly isolated and traced him back direct to
lower mammal forms, it is not to be wondered at that the old master lost
patience. He was no longer capable of controlling developments, or of obey-
ing them; to withdraw from the struggle, which would have been the wisest
thing for him to do, was more than his unbounded energies could endure —
perhaps also he was too vain to do so — and so he continued the struggle
on behalf of his natural philosophy, becoming, as the years went on, more
and more isolated from his old friends and disciples in the world of science.
In compensation he gained from another quarter a new and grateful public.
The old political radicalism had died out towards the close of the century;
most of the liberal party ceased altogether from offering opposition; instead
the struggle was taken up with increasing success against the government
authority by the socialistic labour movement, which, violently persecuted
by Bismarck, sometimes counteracted, sometimes favoured by his successors,
waxed stronger and stronger, until in the revolution of November 191 8 it
destroyed the old social order. With youthful idealism its members embraced
the modern natural science; they too were enemies of the conservative State
Church, which was friendly to the Government and which condemned them
to show humble obedience to superiority. There was all the more reason,
then, for their being drawn together by a natural-scientific explanation of
the world which made progress the aim of life. To them Haeckel's monism
was a welcome ally; that its cosmic view was over-simplified and falsely
depicted it was not in their power to control, owing to their lack of special
studies, but its founder's ardent belief in natural science and intense hatred
of the State Church, combined with his oppositional attitude in politics,
sounded irresistibly attractive. It is against this background that Haeckel's
later scientific activity must be viewed in order that its influence may be
understood aright.
5X4 THE HISTORY OF BIOLOGY
Haeckel' s Weltratsel
In the nineties Haeckel returned to natural philosophy; he published one
or two papers on monism and an important work on "systematic phy-
logeny," comprising a genealogical tree for all living beings — that is, a
detailed application of his earlier, and even then somewhat out-of-date,
theorifes. In 1899 he published his famous work Die Weltratsel, which was
intended to be a summary of his ideas and at the same time a farewell to
his activities; being a child of the nineteenth century, he wished to con-
clude his work with its exit — a promise that unfortunately he failed to
keep. The Kiddle of the Universe had extraordinary success; in Germany the
book was sold by the hundred thousand and in England by tens of thou-
sands; special emphasis has been laid on the fact of its widespread distribu-
tion among the working-classes, and in Japan it is said to have been used
as a school text-book. Nevertheless, from a scientific point of view it must
be regarded as utterly valueless. Its biological section is a rehash of the his-
tory of the creation; anthropogeny, and the monograph on the plastidule,
as little attention as possible being paid to the immense progress made by
scientific research since then. As a matter of fact, biology takes up only one-
quarter of the volume; the rest is devoted to psychology, cosmology, and
theology. The cosmological section gives evidence of the author's hopelessly
confused ideas on the simplest facts of physics and chemistry; final judgment
has been passed on it in a widely distributed polemical paper, which has
never been challenged by trustworthy authorities, written by the Russian
physicist Chwolson, to whom we refer those who desire to gain an insight
into Haeckel's standing in regard to the exact sciences. The philosophical
section of the book has been no less severely criticized by specialists on the
subject; philosophers of different schools have pointed out its utter lack of
clarity in point of theoretical knowledge and logic, its incapacity to define
even the simplest ideas. In passing, it may be mentioned that this time "mon-
ism" is based mostly on Spinoza, the great dogmatist and repudiator of evo-
lution, whose purely metaphysical idea of substance is at once placed on
a par with the "matter" of physics. True, the real character of substance
is said to be inexplicable, but, notwithstanding this, everything between
heaven and earth is explained with its aid. If we add to this Haeckel's total
lack of historical sense and critical judgment — his views on events and
persons are derived from the simplest vocabulary of contemporary political
and cultural radicalism — the final impression of The Kiddle of the Universe
will be an utterly depressing one. The cause of the book's popularity is
obviously to be found in the political and social sphere. Its very introduction
points in that direction, the progress in the scientific world being there con-
trasted with a dark picture of the political situation of the time: government,
administration, courts of justice, and education are depicted as appallingly
MODERN BIOLOGY 5x5
behind the times, and, above all, the Church is, of course, represented as
the centre of all kinds of obscurity, superstition, and tyranny. From all
quarters, both radical and conservative, the signal to open hostilities was
eagerly awaited. Some years after the appearance of The Kiddle of the Utiiverse
there was founded the Monist League, a widely ramified association formed
for the purpose of working for the ideas to which Haeckel gave expression
in this book and in a sequel to it. Die Lebenswunder. Since then it has laboured,
by means of meetings, lectures, and papers, and in some circles by devotional
exercises, with a view to taking the place of the ecclesiastical cult. Haeckel's
colleagues, however, for the most part kept aloof from the league; only a
few scientists of importance have joined it. From the side of the conserva-
tives violent attacks were made on the league; in the Prussian Diet Reincke,
the professor of botany, made a strong stand against it, characterizing it as
a menace to society and subversive of morals. This started the battle in ear-
nest. To counteract the Monist League there was founded the Keplerbund,
so-called after the great astronomer. The very name, however, proved fatal;
Kepler, it is true, was at the same time a great naturalist and a devout Chris-
tian, but all the same he was so saturated with the grossest superstitions of
his time that he cannot by any stretch of the imagination be held up as the
ideal seeker after truth in modern times. And the Keplerbund failed no less
than the Monist League to attract scientists of any weight; the latter kept
more strictly than ever outside the struggle and showed on the whole —
the biologists, at any rate — their sympathy for Haeckel, whose work, in
spite of all his mistakes, nevertheless seemed to them to represent a struggle
for enlightenment and liberty of doctrine against the constant menace of
the powers of reaction.
And Haeckel certainly did maintain the radically liberal-minded stand-
point of his youth undisturbed through all these changes — which it was
all the more easy for him to do as he had never taken part in practical poli-
tics and therefore had not to solve any political or social problems of detail.
But the shock caused him by the Great War proved all the greater on that
account; the idea that the fellow-countrymen of Darwin should have sided
with the enemies of Germany drove him to despair. A few more works came
from his pen, among them one entitled Funf^igjahre Stammesgeschichte, with
which he celebrated the fifty years' jubilee oiGenerelle Morpbologie, and which
testifies to his having learnt nothing and forgotten nothing. His final work,
Krisfall-Seelen, is sufficiently characterized by its title. It came out in 1917; two
years later he died, his death being hastened by an accident, which de)ivered
him from the infirmities of old age and the misery of those unhappy years.
Hartjnann on Haeckel
The well-known philosopher Eduard von Hartmann, in an otherwise sym-
pathetic character-sketch of Haeckel, describes the latter's "monism" thus:
5 26 THE HISTORY OF BIOLOGY
"He is an ontological pluralist in that he conceives nature to be a multi-
plicity of separate substances (atoms), a metaphysical dualist in so far as
he assumes in every substance two combined metaphysical principles (en-
ergy and matter); a phenomenal dualist in that he assumes two distinct
spheres of phenomena (external mechanical happening, and internal sensa-
tion and will), a hylozoist because he ascribes to all matter the possession
of life and soul; further, he is a philosopher of identity, a cosmonomistic
monist, and a materialist." Thus the Haeckelian monism, if closely looked
into, will be found to contain a little of everything. It may therefore be
worth pointing out in this connexion that even more deeply elaborated mo-
nistic systems have appeared in our own day. As a matter of fact, monism as
a philosophical view of life is a comparatively ancient doctrine; the neo-
Platonists, who ascribed true existence only to ideas, were undeniably
monists, as was also Spinoza, and so, too, Schelling and his successors, in-
cluding both Goethe and Hegel. Monism based on natural-scientific grounds,
however, has undoubtedly become an especially widespread conception in
modern times. As one of its leading representatives may be mentioned Ernst
Mach (i 838-191 6), professor originally of physics and then of philosophy
at Vienna. As a physicist he applied himself, inter alia, to mental-physio-
logical studies after the pattern of Helmholtz, but he also studied Kant's
writings and was led through them into the sphere of the theory of knowl-
edge. He thereupon felt himself called upon .to create a method of scientific
thinking, not as a philosopher, for he was unwilling to call himself that,
but as a student of science. He will have nothing to do with transcendent
spheres of thought. Through the analysis of different sense-impressions he
came to the conclusion that everything is phenomenal; nothing exists in
itself; the outer world consists of a series of phenomena, and the ego, the
personality, likewise of a series of phenomena, which we call perceptions;
the phenomena stand in a relation to one another, which is expressed by
the functional terms of mathematics: one change brings about another; the
phenomena inside and outside the personality are mutually interdependent.
Mach denies the principiant contrast between appearance and reality; the
most fantastic dream is just as much a phenomenon as a real event; he like-
wise denies the contrast between ego and non-ego, for both are a series of
mutually interdependent phenomena. The manner in which Mach explains
on these postulates such phenomena as will and thought has been much dis-
cussed by philosophers who have made a special study of the subject, and
has often been characterized as lacking in seriousness: by Hoffding, for in-
stance, who points out that the elements common to physics and physiology
are in Mach indefinite and mystical, like a shapeless nebula. Now, Mach,
as already mentioned, claims to be only a natural scientist and to try to
solve only natural-scientific thought-problems. But even as such he exposes
MODERNBIOLOGY 5x7
himself to the same criticism; his biological reasoning must thus be regarded
as out of date even for his age, and partly also somewhat ingenuous; he
argues about evolution and heredity without taking into account the re-
sults of contemporary research, he believes in much of the old, childish ani-
mal-psychology in the spirit of the earlier Darwinism, and he speculates,
like Haeckel, upon the possibility of explaining the origin of the sense-
organs by means of the theory of selection. Strangely enough, he also de-
fends the teleological explanation of nature, as far as biology is concerned —
though as a provisional explanation only, until a true causal explanation
is forthcoming. His references to all that teleology has achieved in arousing
interest in problems and collecting facts with which to solve them may not
be devoid of truth, but he certainly overlooks the confusion it has caused
by inducing vitalistic explanations of nature; such, in fact, were revived
under the influence of Mach, as we shall see later. Finally, with regard to
his monism, it possesses, in spite of his own assurances, more philosophical
than scientific interest; the practical scientist should at any rate be allowed
to treat things as really existing and the changes that take place in them as
having been causally effected.
Another monistic theory was set up by Richard Avenarius (1843-96),
professor of philosophy at Zurich. He, too, elaborated a kind of theory of
function, but in contrast to Mach he gives to its elements a material nature.
His theory, which suffers from having been presented in very difficult lan-
guage, has had less influence than Mach's.
Natural-philosophical theories of this kind may offer some interest as
thought-experiments and besides may have their ideal value, if they give
expression to the conception of life of a consummate personality. Exact
scientific research, on the other hand, carves out paths of its own, its prog-
ress sometimes hindered, sometimes furthered by the different conceptions
of the world, according to how they deal with existing facts. Pasteur, for
instance, in the controversy over spontaneous generation, undoubtedly de-
rived advantage from his Catholic dogmatism as against those who saw in
spontaneous generation a "philosophical necessity." And his very example
shows, too, how in the long run the practical utility of observations is the
most conclusive criterion of their value. Those facts will last which con-
tribute, however indirectly, towards extending man's dominion over na-
ture, whereas the " theories of life," after surviving for a time, find a haven
in the archives of cultural history, provided they are found worthy to be
preserved there.
CHAPTER XV
MORPHOLOGICAL SPECIALIZED RESEARCH UNDER
THE INFLUENCE OF DARWINISM
1. Anatomy and Embryology
Development of anatomy
THERE IS NO DOUBT that the power of Darwinism reached its zenith
in the seventies and eighties. By then the opponents of the earlier
school had for the most part said their last word, and the younger
generation of scientists who had embraced the new doctrine as yet found
no difficulties in its application. Rather, efforts were made, by means of
exhaustive investigations in every possible field, to collect fresh proof for
it. These endeavours resulted in an extraordinarily abundant and many-sided
production, chiefly in the sphere of morphology, with its various special
subjects, though also in those of geography and oecology, as well as in the
purely systematic sphere. In this chapter we shall give a comprehensive re-
view of this specialized morphological research -work, which was as many-
sided as it was rich in results.
Anatomy developed as the outcome of a number of investigations, the
results of which were recorded in numerous memoirs. To give an account
of all the valuable facts that were brought to light in the course of this
ceaseless work would be impracticable within a reasonable compass; a mere
list of the anatomical works that were published during that period would
run into hundreds of pages. In the field of the invertebrates especially, in-
numerable new and important anatomical discoveries were made; hitherto
unknown, or at least neglected, animal forms were now studied and often
produced undreamt-of ideas for the furtherance of comparative research.
Chastognatha and Enteropneusta, Tunicata and Brachiopoda may be men-
tioned as examples of such forms, which, though insignificant in their ap-
pearance and scope, are nevertheless interesting for their structure and
development. But the Vertebrata also continued to provide valuable con-
tributions to comparative anatomy, which, for the very reason of its mor-
phogenetical aims, found every animal form, however insignificant, worth
while investigating and examining for the circumstances of its origin and
evolution. But, on the other hand, by reason of the aims they had in view,
518
MODERN BIOLOGY 5^9
these investigations ultimately became somewhat monotonous, with the re-
sult that this line of research finally became quite unmodern and the interest
began to turn in other directions. Among the investigators who compiled
the results of this work may be mentioned Robert Wiedersheim (1848-1913),
a disciple of Leydig and professor at Freiburg, well known for his compre-
hensive work on the anatomy of the Vertebrata as well as his studies of
special subjects, particularly of the bone-structure of the Batrachia, and the
Swiss, Arnold Lang (1855-1916), a disciple of Haeckel and professor at
Zurich, who wrote a widely referred-to work on the anatomy of the in-
vertebrates and a number of monographs on various groups among the
worms,
Etnbryology
That branch of morphology, however, that was specially developed under
the influence of the descent theory was embryology. The biogenetical prin-
ciple and its related subjects, the theories of germinal layers and the gastrasa,
were applied to different spheres and gave rise to ideas in many directions,
besides which the new microtechnics offered a means for detailed discover-
ies of hitherto undreamt-of results. Embryology, therefore, proves to have
been the most productive of the morphogenetical specialized spheres and
attracted to it the most eminent biologists of the time.
Among these representatives of phylogenetical embryology only a few
of the more important can be mentioned here. Alexander Kowalewsky
(1844-1901), an academician of St. Petersburg, worked in the spirit of
Haeckel, encouraged by his commendation; his detailed investigations into
the development of ascidians and salpa; covered an immense amount of de-
tail and the same is true of his work on the development of the lancet-fish,
with the result that even the ontogeny of this much-discussed animal be-
came known. Kowalewsky was a firm supporter of the theory of the ger-
minal layers and developed it by making contributions of his own in the
theoretical sphere.
The same line of research was also followed by the two brothers Hert-
wig, and it led them both to make discoveries of fundamental importance
and to produce theoretical ideas of a very different nature from those from
which they had started. Oscar Hertwig was born in 1849 and Richard
Hertwig in 1850, the sons of a merchant at Friedberg in Hesse. They both
studied at Jena under Haeckel and became lecturers there and finally pro-
fessors, Oscar of anatomy at Berlin, Richard of zoology at Munich. Both
carried on, each in his own subject, extensive and important activities as
teachers and investigators. At Jena they worked together in the sphere of
evolution in the manner of Haeckel and published a series of papers entitled
Studien x.ur Blattertbeone, which dealt especially with the problem of the mid-
dle germinal layers. Here they expounded their famous "coelom" theory,
530 THE HISTORY OF BIOLOGY
which was intended to be a universal answer to the question: "How does
the two-layered embryo develop into a higher organization?" The theory
takes as its starting-point the two primary germinal layers, the ectoderm
and the entoderm, between which there arises at an early stage an originally
structureless layer formed by immigrating cells, which is here termed "mes-
enchyme." The animals are now divided, in respect of their development,
into two groups, dependent upon whether the mesenchyme participates in
the formation of tissue or not. The former takes place chiefly in the coral
animals, the flat-worms and the molluscs, in which the muscular and nerv-
ous systems are formed out of the mesenchyme, whereas in most other ani-
mal types, chiefly the Articulata and the Vertebrata, the said tissues are of
purely epithelial origin and are formed out of a dual evagination of the
entoderm, the inner cavity of which gives rise to the body cavity, or the
coelom. The theory was afterwards applied, after a series of special investi-
gations, to the organic formation of different animal forms and won general
acceptance at the time. It is true, His declined to accept it, but did not suc-
ceed in substituting any better explanation. Later research, however, has
found this theory to be far too schematical; students have given up referring
the various organs to the three germinal layers and now instead seek their
origin, each separately, in so-called primitive rudiments. Furthermore, the
formation of the coelom through simple invagination has been found upon
closer investigation to be far less frequent than the two brothers imagined.
Their theory has nevertheless played its important part and has called forth
abundant special research-work of value for all time. In the following pages
we shall repeatedly find their names mentioned in connexion with valuable
contributions to the advancement of biology. Among their pupils may be
cited the scientific collaborators Eugen Korschelt (born in 1858, professor
at Marburg) and Karl Heider (born in 1856, latterly professor at Berlin),
who together published an exhaustive summary of the knowledge of their
time regarding the evolution of the invertebrates. Moreover, both have dis-
tinguished themselves as specialists, particularly in the sphere of experimen-
tal research.
During this period England was also the scene of valuable embryologi-
cal research-work. Among her representatives may be mentioned Edwin
Ray Lankester (born 1847), a professor at the British Museum and author
of a number of papers on evolution, dealing especially with the fishes and
the Articulata. He especially took up for study and further elaborated the
coelom theory and has brought it to the highest point it has yet reached,
having sought to base on it the classification of the animal kingdom. A
very distinguished name in the sphere of evolution has been won by Francis
Maitland Balfour, who was born in 1851 and died, as the result of an
accident, in i88i, the younger brother of the famous statesman Lord Balfour.
MODERN BIOLOGY 531
During his short life he found time to carry out a number of extremely
important works on evokition, including a study of the evolution of the
sharks and a Treatise of Comparative Embryology, giving an account of the evo-
lution of the Q.gg and the embryo throughout the animal kingdom, a work
that was of unrivalled importance at the time; an application of modern
genetical embryology to the whole animal kingdom and at the same time
a powerful defence of the Darwinian morphology in its classical form. Bal-
four, in fact, definitely maintains that phylogeny is the goal of evolution,
while at the same time in certain details, as, for instance, in the theory of
extremity-formation previously mentioned, he adopts a dissentient attitude
towards the contemporary Gegenbaur school.
Even by then the morphogenetical embryology had met with decided
opposition on the part of the naturalists who desired to substitute for phy-
logenetical conclusions the study of function in those organs whose evolu-
tion was under investigation, and thus to give evolution a more or less
physiological direction. To this group belongs the afore-mentioned Wil-
HELM His (1831-1904), who was born at Basel, became professor of anatomy
there, but was afterwards summoned to Leipzig, where he worked until his
death. Famous both as an anatomist and as an embryologist, he paved the
way for a new line of research, particularly in the field of embryology. First
of all he expected to see in the evolution of the embryo a physiological
process, the course of which should be so studied that each later stage of
development must necessarily proceed from the immediately preceding one.
The changes whereby the simple egg-cells are formed into complex organisms
are, to his mind, purely mechanical; as the result of a series of flexions,
fold-formations, and accretions the embryo arises out of the originally
lamellate germinal layers, and its folds are in their turn produced entirely
from variformed growth. Every organ possesses its given rudiments in the
germinal layers and these layers thus consist of a quantity of " organbildende
Keimbex.irke" ; they are therefore not indifferent, as C. F. Wolff and, after
him, Haeckel declared. In connexion herewith His sharply criticizes the
biogenetical principle; embryos of different animal forms are as easily dis-
tinguishable from one another as the fully developed animals; Haeckel 's
proofs to the contrary, both verbal and pictorial, are examined and found
to be untenable, and finally the question is put: "If we possessed a complete
genealogical tree, would our own or any other extant organic form be fully
explained thereby?" In reply His declares that if in a case of near-sightedness
it is possible to establish the fact that the individual in question has inherited
the defect, little will have been gained therefrom as regards our knowledge
of the character of that defect; rather, the eye's capacity for accommodation
and other concomitant circumstances must be investigated for this purpose;
in the same way, the physiological side of embryonic development is more
53X THE HISTORY OF BIOLOGY
important than any phylogenetical speculation. The whole of this way of
thinking won but little acceptance in his own time, when research was being
directed along phylogenetical lines; in the eyes of posterity, on the other
hand. His stands out as precursor of the mechanical method of evolution,
which has since won so many adherents and which will be dealt with in
the following pages.
His, however, was by no means alone, even in his own age, in his con-
ception of evolution. There were others who also opposed the one-sided phy-
logenetical line of research; among them may be mentioned Alexander
WiLHELM GoETTE (1840-19x1). He was born at St. Petersburg of Baltic ori-
gin, studied at Dorpat and at Gottingen, and finally became professor at
the German university in Strassburg, where he worked for the greater part
of his life. As an embryologist he was influenced from the beginning by his
fellow-countryman von Baer. In his principal work. Die Entwkkelungsge-
schichte der Unke, he seeks to make the evolution of Bombinator igneus the basis
of a. purely mechanical theory of evolution, freed from both Haeckel's " jorm-
bildende Krdffe" and Gegenbaur's phylogenetical constructions. Starting from
the old, but at the time commonly accepted, delusion that the nucleus of
the egg is dissolved before fertilization, he declares that the egg is an "un-
organized, inanimate mass," wherein are formed by purely mechanical forces
— osmotic currents and resultant pressure-changes — the first divisional fur-
rows, and together with them fresh nuclei as centres for the development
of the new cells. Thus is explained the origin of life out of lifeless substance.
Similar mechanical explanations are then invented for the formation of the
germinal courses and organs. In a later work Goette deals in the same method
with the stages of development in certain worm-forms. The interest attaching
to these investigations lies in the mechanical conception of the embryonic
development, which is not only maintained theoretically, but is also in
many respects successfully applied. Unfortunately, Goette was so delighted
with his theory that he let all criticism go by the board; his false concep-
tion of the nature of the egg he still maintained long after it had been proved
untenable; his detailed research was extremely arbitrary and was severely
criticized by Gegenbaur. His theory has nevertheless not been without its
effect; his disciple Roux especially, doubtless under his influence, formu-
lated a mechanistic conception of embryonic and organic development that
received widespread support.
Another opponent of the universally current embryological conception
was NicoLAus Kleinenberg, born in i84x at Mittau, a disciple of Haeckel's,
and eventually professor at Palermo, where he died in 1897. Of his literary
production, which was small in extent but original in character, may be
mentioned his monographs on the evolution of the fresh-water polypus, in
which the ontogeny of this primitive animal is elucidated for the first time.
MODERN BIOLOGY 533
In a subsequent work on the course of development of an annelid he has
propounded a curious theory of its embryonic evolution in strong opposition
to Haeckel's gastrxa. theory and the coelom doctrine of the Hertwigs. He
starts with the sentence: " Es gibt kein tnittleres Keiinblatt," and adduces a
number of examples of how various organs that had been supposed to be
mesodermal, originate directly or indirectly from the ectoderm or entoderm.
At the same time he maintains that the form of one organ depends upon its
function and not upon its origin; "A comprehensive tissue-system is pos-
sible only on a physiological basis." These views were at the time at which
they were expressed (1886) so utterly opposed to those of his age that they
scarcely caused any sensation; their time came later, in connexion with the
altered view of evolution that has become prevalent in our day, which will
be described in a subsequent chapter.
z. Cytology
Development of microscopical cell-research
Microscopical cell-research is undoubtedly the branch of biology that re-
ceived the greatest stimulus during the last decade of the past century, and
that has seen the most important results and in many ways set its mark
upon the whole of biology in general. Its highly perfected methodics, with
its minute technical preparation of material for investigation, carefully
adapted to suit each particular case, and its careful microscopical study of
the smallest details, employing the highest possible magnifications, became
a characteristic feature of the research work of that period. The purely tech-
nical side of biology thereby received an entirely new character; whereas
formerly skill in dissection was the most essential qualification of the biolo-
gist, this ability now became to a certain extent superfluous, thanks to the
development of the technique of microtomy. On the other hand, the student
of cells, if he desires to create something new and to work independently,
must acquire a chemical knowledge of the means of fixing the tissues, as
well as a colour technique for the purpose of their further treatment. It was,
of course, possible for the whole of this method of research to degenerate
into a mere unintelligent dexterity, as biologists of the old school in par-
ticular called it, but it has also made possible more than any other method
the obtaining of results that have entirely transformed our conception of
the phenomena of life.
We left cell research at the point to which Max Schultze had brought v
it — the knowledge of the cell as a limited quantity of protoplasm with
concomitant nucleus. Schultze is also remarkable inasmuch as in his cell
studies he was still working without a microtome; he brought cytology to
534 THE HISTORY OF BIOLOGY
the farthest point possible with the old methods. During the period imme-
diately succeeding his death, cell research made rapid strides in regard both
to the value of the discoveries made and to the number of workers engaged
in it. Limitations of space make it impossible to do justice to all the truly-
distinguished minds that during this period laboured for an increased knowl-
edge of the life and structure of the cell. Some of the most prominent cytolo-
gists will be mentioned here, after which a description will be given of the
most important discoveries in their field of research; the fact that their activi-
ties and the rivalry to achieve the most important results were contempo-
raneous would indeed render it extremely difficult to retain here the bio-
graphical method of presentation of the subject that we have followed
hitherto.
Its representatives
In the sphere of plant cytology Eduard Strasburger takes the first place.
He was born in 1844 of German parents at Warsaw, received his school edu-
cation there, and studied partly in Paris and partly at German academies,
first at Bonn and then at Jena, where Haeckel won him over to Darwinism
and even procured him a professorship. He afterwards became professor at
Bonn, where he worked until his death, in 1911. Equally distinguished as
a research-worker and a teacher, he attracted to his institute a large num-
ber of pupils from all countries; he was a leading writer of text-books, and
his scientific production included, besides his epoch-making cell-studies, a
number of branches of vegetable anatomy.
Among students of animal cytology the above-mentioned brothers Hert-
wig take high rank; besides them there is Walther Flemming (1843-1905),
professor first at Prague, then at Kiel, distinguished not only as an observer,
Ijut also as a technician and teacher. Further, Hermann Fol (i845-9x); a
native of Geneva and the son of wealthy parents, he studied in Berlin and
became professor in his native town and a scientist of high repute. Being
specially interested in marine research, he equipped at his own expense a
vessel for the purpose; in the course of a voyage, he, together with the ves-
sel and the crew, disappeared and were never heard of again. Otto Butschli
(1848-1910), after having studied chemistry and mineralogy, devoted him-
self to zoology and became professor at Heidelberg. It is possible that his
earlier occupying himself with inorganic elements and processes induced in
him that liking for comparison between organic and inorganic structures
which characterized his later research work. Besides these names should be
mentioned that of the Belgian Edouard van Beneden (1845-1910); the son
of a highly reputed zoologist, who was especially known as an expert on
parasitology, he applied himself to the study of medicine and eventually
became a professor at Liege, famous as a many-sided investigator and pub-
lisher of the well-known journal Archives de biologie. Finally, reference should
MODERN BIOLOGY 535
be made to Theodor Boveri (1861-1915), a disciple of the brothers Hertwig
and professor at Wiirzburg, as well as the two Heidenhains — Rudolf
(1834-97), a pupil of Ludwig and professor of physiology at Breslau, but
active also as a cytologist, and his son Martin, born in 1862. and professor
at Tubingen, who devoted himself exclusively to cell research. All the above
have advanced their science by making valuable discoveries and important
technical improvements.
Strasburger on the formation and division of cells
In 1875 "^^^ published the first edition of Strasburger's pioneer work Zell-
bildung und Zellteilung, a third completely revised edition came out in 1880.
The main problem that occupied cytological research during this period was
that of the origin of the cellular nucleus. As we have seen, Nageli had al-
ready observed the division of the nucleus, but neither his own nor other
similarly extensive observations were able to possess general application.
Even in the first edition of his said work Strasburger makes the nucleus of
the egg-cell in the plants he investigated dissolve upon fertilization and its
mass disperse into the plasm of the cell; in the latter are then formed a
number of concretions, which give rise to fresh nuclei. In the third edition,
on the other hand, it is asserted that examples of independent cell-formation
can no longer be cited from the vegetable kingdom; fresh nuclei invariably
arise through the division of older ones. This established one more of the
principles of modern cytology. Even before this students had begun to ob-
serve the curious phenomena attending nuclear division in the majority of
cells, but apart from these scattered observations, it was Strasburger who,
as far as the vegetable kingdom is concerned, elucidated this process, which,
though complicated, is now widely known and is set forth in all text-books.
This process — indirect nuclear division, also called "mitosis" or "kary-
okinesis" — is as follows: the nucleus, having lost its membrane, concen-
trates its colourable contents around its middle plane, after which the latter
divides itself and the two halves go each its own way and thereupon again
concentrate into two daughter-nuclei. The main principles of this process
were already elucidated in the above-mentioned first edition of Strasburger's
book, and in the third a number of further details are given. In the field of
zoology Biitschli, O. Hertwig, and Flemming during the same decade made
their decisive contributions to our knowledge of nuclear division, and, be-
sides, certain isolated details were discovered by Fol, van Beneden, and
others. As a result of this research work the elements composing the nucleus
were also investigated; filament substance and nucleolus, nuclear juice, and
nuclear membrane were the constituents that were distinguished to begin
with. Of these the first-mentioned was, owing to the part it plays in the
nuclear divisions, the object of greatest attention, and especially on this
subject Flemming's studies of the cells of amphibious larva; were conclusive.
536 THE HISTORY OF BIOLOGY
It was he who ascertained after detailed study the process of nuclear
division and actually gave to its various phenomena the names that have
been in use since then; owing to its strong colourability he called the fila-
ment substance "chromatin" and non-colourable substance, which also ap-
pears upon division, "achromatin"; the names of the different phases of
division — spirem, aster, metakinesis, dyaster, and dispirem — were also
invented by him. He also proved the conversion of the chromatin from a
network into a convoluted filament and further into a number of bent staves,
and proved that the actual division consists in these latter's splitting along
their length. These chromatin staves were afterwards called by Waldeyer
"chromosomes" and have, as is well known, come to play a decisive part
in modern heredity-research. The processes of the fusiform achromatin fila-
ments during division were also studied by Flemming; it was not until later
that the minute central body, the centrosome, which is of such vital impor-
tance for their transformation, was investigated, primarily by Flemming and
Boveri, who together with van Beneden discovered its division in cell-
reproduction. Boveri's studies of the centrosomes especially were very in-
tensive and have proved to be of fundamental importance. This formation
has been characterized by him as the cell's dynamic centre, which facilitates
the nuclear and cellular division. He also discovered that the centre of the
spermatozoon is formed thereby.
While the nucleus has been found in the difi'erent forms of life to repre-
sent the conservative element — the chromosomes are, as is well known,
equally numerous and similarly formed in all cells in the same individual —
the protoplasm of the cell and its many and various derivatives have ofi'ered
fresh problems, owing to their wealth of form, which has proved all the
greater, the more these formations have been investigated. The actual basic
substance, which still has to bear the clumsy and illogical name of pro-
toplasm, has been investigated by a vast number of students and has called
forth many attempts at an interpretation of its essence. These attempts have
for the most part concentrated upon three diff"erent theories based on obser-
vation, which have been named after their founders: Biitschli's froth theory,
Flemming's filament theory, and Altmann's granule theory; to say nothing
of the purely speculative attempts to discover the fundamental substance of
life. The chief difficulty that revealed itself in these explanations and that
brought out their mutual contradictions is actually caused by the incon-
stancy which the living protoplasm always displays and which is a neces-
sary consequence of its role as bearer of all the metabolism in the cells and
the organisms composed of them. Even the nucleus displays phenomena of
substance-renewal and it has been found that the vital manifestations of
the cell ultimately receive their impulses from that quarter, but it is in any
case in the plasma substance that these manifestations of life are essentially
MODERN BIOLOGY 537
expressed. And they are as much of a chemical as of a physical nature; the
physical phenomena — movement of various kinds — are invariably in-
duced and brought about by chemical reactions and in their turn produce
new reactions.
Plasma ttoeories
BiJTscHLi's froth theory is an essentially physical attempt to explain the
structure of protoplasm. He certainly repeatedly points out the chemical
reactionary phenomena of the cell, but he pays little attention to them.
Taking as his basis the strongly vacuolized substance of the lowest pro-
tozoa, especially of the amoeba;, with the current-phenomena visible therein,
he conceives the living protoplasm as a fluid mass identical in its structure
with the emulsion that is obtained when oil and soda-solution are shaken
together. This purely mechanical emulsion-theory he afterwards elaborated
after making a series of experiments of a very ingenious character. Through
the mixture of variously composed liquids both he and a whole school of
investigators after him succeeded in imitating in a surprisingly natural way
a great many of the most complicated movements and structures of the liv-
ing cell-substance. It cannot, of course, be denied that the mechanical phe-
nomena which were found in these experiments to cause the movements in
the given substratum may also be capable of asserting their influence upon
the plasma movements, but as a reproduction of the phenomena of life these
experiments possess the fundamental fault of entirely disregarding the chem-
ical reaction that is incessantly going on in living substance; the mobile
oil-emulsion remains chemically what it was, whereas a creeping amoeba
is continually changing its chemical composition, so that movement and
chemical reaction are indissolubly dependent upon each other. In connex-
ion herewith we find also the belief, which has proved unsatisfactory from
the very beginning, that the fundamental substance of life is fluid — a the-
ory that has been considerably revised by modern colloid chemistry, of which
we shall have more to say presently.
Flemming's plasma theory undeniably takes more account of chemical
conditions. According to this theory, protoplasm consists of a network of
fibres embedded in a homogeneous substance. These structures he found par-
ticularly in the cellular mass in various tissue-elements: in egg-cells and in
cartilaginous and glandular cells in higher animals. He believes the phenom-
ena of metabolism in the cell to be accompanied by changes in the filament
mass and in the basic substance, which should be examined in detail in
different subjects. The threads may sometimes be dissolved into canals and
vacuoles and thereby convey the assimilation products not only to different
places within the cell, but also between various cells, for these latter are
in most cases demonstrably connected with one another by bridges of fila-
ments. Thus the cells become the structural elements in the body, though
538 THE HISTORY OF BIOLOGY
also elements incorporated in one and the same vital unit; their independence
need not be over-stressed.
In opposition to these two theories, which belong to the eighties,
there appeared somewhat later the granule theory of Altmann. Richard
Altmann (1851-1901), professor at Leipzig, devoted his attention chiefly
to the fundamental substance in which the above-described network of
plasm lies embedded, and with the aid of suitable colouring-matter he found
in it a mass of grainlike formations — the Latin gramda — of different kinds
in different cells. In these he sees the true substance of the cell; he even finds
that the threadlike structures which can be produced- by Flemming's method
are composed of similar granular formations. Indeed, many of his observa-
tions have been confirmed; in the glandular cells especially, the forthcoming
secretion first appears in the form of homogeneous granules, which gradu-
ally increase in size and assume the form of drops. In most other cells, too,
such granular structures appear as expressions of the cell's change of sub-
stance; as in nerve- and muscle-cells, of which we shall have more to say
later. Altmann, however, sees far more than this in these granule formations;
he calls them bioblasts and considers them to be the true elementary organ-
isms of which cells and tissues are composed, just as bacterial colonies are
composed of various bacteria. He even believes these protoplasmic granules
to be of equal value to micro-organisms and would make this his contri-
bution towards the solution of the riddle of life, a contribution that he
further supplements by finding a resemblance between bioblast and crystal;
these two are in fact compared, though hypothetically. These fantastic ideas
have naturally been given but little support; Altmann is on firmer ground,
however, when he emphatically states that the living substance must be
solid and not liquid — an assertion that he bases upon his granule theory
in opposition to Biitschli's above-mentioned experiments and speculations.
These granular structures of the cell-substance have, as a matter of fact,
been studied by numerous later investigators, who have given them innu-
merable names: "mitochondria," "chondriosomes," etc. They are brought
to light by the use of special colouring-methods, but in favourable circum-
stances they may also be visible in the living subject, which justifies the
assumption that they are not pure artificial products. The same, indeed, is
true of the other two plasmic structures: the fibre and the froth structures.
M. Heidenhain rightly points out the possibility of all three structural forms'
existing in one and the same cell. But this would also go to show that none
of the structural theories is capable of forming the basis for a uniform con-
ception of what living matter is really composed of. Heidenhain therefore
holds that the common structure of the living plasm must be sought beyond
what is microscopically visible — that it consists in a system of minute
particles that possess the qualities of life, principally that of multiplication
MODERN BIOLOGY 539
by fission, and that build up those structures of which the cell is composed.
These ultra-microscopical particles, which therefore cannot be observed,
but the assumption of which he considers to be an incontrovertible neces-
sity, he calls "plasomes." It is not worth while going further into his spec-
ulations; it will at once be realized that we here have a name for Haeckel's
plastidules, Darwin's gemmules, and innumerable similar ideas — unknown
quantities that can be used neither for the purposes of observation nor for
theoretical calculation and that are therefore automatically eliminated from
the problem of life. True, ultra-microscopical technics have since given us
some insight into the composition of the living substance over and above
w^hat the microscope has been able to provide, but no one has succeeded in
isolating any vital unit in this way, and up till now the cell, with all its
complications, remains the smallest form under which the living substance
has been found to exist by itself and independently of other living entities.
Of undoubtedly greater value have been those facts in regard to the composi-
tion of the cell that have been contributed by modern chemical research,
which will be discussed later.
While, then, the fundamental substance of the cell has remained in its
innermost essence undiscovered, careful and extensive studies have been de-
voted to the mass of cell products of which the bodily tissues are built up.
Of the pioneer research-work in this field may be mentioned the investi-
gations of the elder Heidenhain into the glandular secretion in man and the
higher animals, as a result of which light has been thrown for the first time
especially upon the microscopical structure of the salivary glands and the
relation between the composition of their cells and the nature of their se-
cretion. In his footsteps followed Flemming, Altmann, the younger Heiden-
hain, and many other cytologists, who observed and compared the different
phenomena in the epithelial cells, both the secreting and the resorbing, in
the various organs of the body and the cells covering the surface of the body.
In this sphere the study of the origin and development of the granular forma-
tions has been most intensive.
Nerve investigations
One field of inquiry that has especially occupied the attention of modern
cytology is the nervous system. Its highly complicated structure long re-
sisted all attempts at an explanation, until methods were discovered whereby
it is possible to colour only certain special elements, which can thus be ex-
amined in their entire length. These "elective" methods include impreg-
nation with metallic salts, which has been applied in various forms by the
Italian Camillo Golgi (1844-19x6), professor at Pavia, the Spaniard San-
tiago Ramon y Cajal, born in 1851, professor at Madrid and an unusally
thorough expert in the elements of the nervous system throughout the ani-
mal kingdom, the author of a number of papers on the subject, as well as
540 THE HISTORY OF BIOLOGY
the monumental work Histologic du sysfeme nerveux, and the Hungarian Stefan
Apathy (1863-1913), professor at Koloszvar. Another method that was pro-
ductive in this respect has been the incravital methylene blue-dyeing method,
which was discovered by Paul Ehrlich (1851-1915), disciple of Koch and
principal of the laboratory of hygiene at Frankfurt am Main, and which
has been further applied especially by A. Bethe, professor at Strassburg.
Others who have studied the nervous system include the aged Kolliker, the
Frenchman Louis Antoine Ranvier (1835-1911), professor at Paris and
active worker in many branches of cytology, and also Gustaf Retzius (1841-
1919), son of the above-mentioned anthropologist and early in life an assid-
uous worker in this sphere. Neurological research has to a certain extent
sought to ascertain the structure of the actual nerve-cells and their internal
modifications during different stages of activity; as expressions for the phys-
iological condition in the protoplasm of these cells have been characterized
the granular formations amassed in stages of rest and disappearing upon
irritation, which are called tigroid substance, owing to their appearance,
or "Nissl's granules" after their discoverer, Friedrich Nissl, hospital doc-
tor at Frankfurt am Main (died 1919). Still greater interest, however, has
been devoted to the problem of the connexion between the nerve elements,
which indeed is of vast importance also from the physiological point of
view. In this field there have been two mutually opposed theories. Even
His had observed that there grow out from the embryonic nerve-cells threads,
which become longer and longer. Later on, Kolliker, Cajal, and Retzius,
among others, held the view that these threads give rise to the nervous
fibrillar and that the nervous system is thus formed of a number of mutually
independent elements, consisting of a cell with its concomitant nerve-thread
and connected with its neighbours only by contact. In opposition to this
view. Apathy in particular has maintained that the nerve-thread is formed
of a whole series of cells and that its ramifications extend not only up to,
but also into, the plasm in the ganglion-cells. The conflict between these
two lines of thought was at one time quite lively, but apparently died down
without either party's being able to claim a decisive victory.
Muscle investigations
Besides the nervous system, the musculature early attracted the attention
of the cytologists, especially the cross-striated musculature, the complex
structure of which had long withstood all attempts to interpret it. William
Bowman (1816-91), professor of physiology in London, was the first to
make any weighty contribution towards the solution of the problem. In a
treatise printed in 1840 he describes how the muscle is composed of fibrillar,
surrounded by a substance that he calls sarcolemma, and how the fibrillar
are divided crosswise into laminx of various degrees of density. During the
time that has elapsed since then, muscular histology has had many students.
MODERN BIOLOGY 541
The most important progress is coupled with the names of Alexander Rol-
LETT (1834-1903) and Theodor Wilhelm Engelmann (1843-1909), both
professors of physiology, the former at Graz and the latter first at Utrecht
and afterwards in Berlin. Both of them have done service in ascertaining
the regular sequence of the cross-stripes in the muscles. Rollett is responsi-
ble for these formations' being denoted, as they still are, by letters. Engel-
mann, a disciple of Gegenbaur and a distinguished investigator in many fields
of research, made a special study of the physical qualities of muscle — the
condition of the various elements in normal and polarized light, upon con-
traction and relaxation. These results led to a one-sided physical view of
muscular action, which was still further advanced by Helmholtz's and other
physiologists' investigations into the mechanics of muscular action. On the
other hand, Emil Holmgren (i866-i9ix), professor of histology at Stock-
holm, Sweden, held a more morphological conception of the muscular
process; by careful experimental and microscopical studies of the granular
formations which, thanks mostly to G. Retzius, were already known, which
are situated between the cross-sections of the various fibrills, he discovered
that the granules are the organs which bring about the change of substance
in the muscle during action; his views were accepted and elaborated by
AuGusTE Prenant, professor at Paris and well known as an unusually many-
sided cytologist and author of that both extensive and intensive work en-
titled Traite de cytologie. We can deal only briefly with the various categories
of supporting tissue — connective tissue, cartilage, bone. In this field of re-
search a number of investigations, important from the point of view of prin-
ciple and masterly in their technique, have been carried out by, inter alia,
Ranvier, Flemming, Studnicka, and the Dane F. C. C. Hansen; these have
discovered especially the origin of the categories of supporting tissues and
their transitions into one another.^
Discovery of fertilisation
Undoubtedly the greatest service to biology that has been performed by
modern cell-research, however, is its having given us our present knowledge
of the course and significance of fertilization — a discovery worthy to be
placed by the side of the explanation of the circulation of the blood in the
seventeenth century. If, however, we compare the course of these two great
achievements in the field of research, we get a striking impression of the con-
trast that exists between scientific activities nowadays and those of a couple
of centuries ago. On the one hand, Harvey, who spent twenty years or so
quietly and peacefully examining the idea that had been kindled in him in
his youth, and who afterwards submits it to the world in its perfect and
^ Accounts of the development of cell research in more modern times will be found in
Prenant's above-mentioned work, in M. Heidenhain's Plasma und Zelk, and in other histological
text-books, to which the reader is referred.
5 42. THE HISTORY OF BIOLOGY
consummate form. On the other hand, a number of scientists of today,
some of them admittedly of the highest rank, working all at the same time
in mutual rivalry at the problem of fertilization, expound their results
practically every year, often in a half-finished state and sadly in need of
emendation, and then disputing each other's claim to the honour of hav-
ing produced the various details of the discovery, each one seeking to inter-
pret according to his own lights statements that are often found to have
been originally formulated as mere assumptions and suggestions. The account
of how our knowledge of the origin of individual life was finally acquired
can therefore hardly be so attractive a task as that of describing Harvey's
lifework.
That science had so long to wait before the phenomena of fertilization
were fully elucidated is, of course, primarily due to the fact that the knowl-
edge of its basis, the cell, was for so long incomplete. And in particular the
idea as to the nature of the nucleus of the cell was, as O. Hertwig so weightily
observes, still extremely vague as late as in the seventies; a cystic, homogene-
ous formation was seen in the middle of the cell, and no really clear idea was
obtained as to its meaning. It was supposed to have been observed that on
certain occasions this formation disappeared and that this was particularly
so within the egg-cell; moreover, it was postulated by Haeckel's biogeneti-
cal principle, according to which every living being arises out of an entirely
undifferentiated mass of plasm. It was thought to be probable that sperma-
tozoa, one or several, penetrate the egg upon fertilization, but the part they
played in the process was utterly vague; on the whole, it was deemed suffi-
cient to assume some kind of chemical or physical influence upon the egg-
cell, whereby its stages of segmentation, which had already been studied,
were produced. Indeed, the phenomena of nuclear division, referred to above,
had been partially investigated in the case of egg-cells; Fol particularly had
observed radial phenomena accompanying division, and Biitschli the actual
nuclear pole, but no one had as yet gained any clear idea of the process.
In 1875 O. Hertwig spent some months by the Mediterranean Sea and
there discovered an object particularly suitable for studies in fertilization
and egg-development in the sea-urchin, whose eggs are transparent, occur
in large numbers, and are rapidly developed. The results he obtained from
his investigations of this material he recorded in a dissertation written for
the purpose of obtaining a lectureship at Jena. Among the theses that ac-
companied the paper, according to the German custom, the first runs as
follows : ' ' Die Befruchtung beruht auj der Verschmelzung von geschlectlkh differen-
Xierten Zellkernen." This statement, upon which further light is thrown in the
paper itself, really contains the essence of our modern theory of fertilization.
There is indeed still another of these theses that is of importance: the as-
sertion that the egg does not pass through any ' ' monera ' ' stage — a statement
MODERNBIOLOGY 5 43
directly conflicting with the master's (Haeckel's) theory. Otherwise this
work is in many parts somewhat incomplete; thus, the so-called pro-
nucleus in the egg is supposed to be developed out of the germinal spot in
the ovarial egg, while the rest of the germinal vesicle is assumed to have
disappeared — this being stated to apply to the entire animal kingdom. No
polar bodies have been observed, nor has it been possible to prove the pene-
tration of the spermatozoa into the egg. On the other hand, what is described
and illustrated is how two nuclei in the egg gradually approach one another
and coalesce, one of which comes from the extreme part of the egg-cell and
is therefore characterized as the nucleus of the male sexual product. As a
matter of fact, Biitschli and others had previously observed two nuclei unite
in the fertilized egg, but had not utilized their discovery for the purpose of
a general interpretation; simultaneously with O. Hertwig, van Beneden had
published an account of certain fertilization-phenomena in the egg of the
Mammalia and had therein expressed the view that, of the two nuclei which
he also had observed in the newly-fertilized egg, one is of male and the
other of female origin, but he made this statement under reserve as being
only a hypothesis, "which may be accepted or rejected." The principle
that fertilization consists in the union of the male and the female nuclei was
thus without any doubt first set forth by Oscar Hertwig; he was the first to
realize the significance of the phenomena and he therefore deserves all the
honour for it.
Our knowledge of fertilization thus made slow progress, with the col-
laboration of different investigators. Fol was the first who actually saw
(1879) the spermatozoon penetrate the egg, thereby establishing what
O. Hertwig had already concluded, that one single male cell performs the
act of fertilization. The latter scientist followed up his studies of fertiliza-
tion and gradually succeeded in arriving at a clearer view of the subject; in
a work on the fertilization of the worms he gives an account particularly of
the expulsion of the polar bodies; these bodies, which have already been
described by Sven Loven — and possibly still earlier by the aged Carus —
were now found to arise through indirect nuclear division, but were still
regarded by Hertwig as one of the more incidental phenomena of fertiliza-
tion. It was not until later that he discovered them in his first subject of
investigation, the egg of the sea-urchin. The next great step towards a solu-
tion of the riddle of fertilization was taken by Flemming, who in 1879
established the longitudinal cleavage of the chromosomes in indirect cell-
division, which was afterwards confirmed by Retzius and Strasburger. In
1887 van Beneden published the results of investigation into the fertiliza-
tion of the lumbrical ascarid worm of the horse, Ascaris megalocephala, well
known on account of its few but large chromosomes. In this animal he found,
and afterwards established in other quarters also, the important fact that
544 THE HISTORY OF BIOLOGY
every animal has an equal number of chromosomes in each cell. In connexion
therewith he discovered the reduction in the number of chromosomes in the
sexual cells: that upon the latter's maturation-division the number of chro-
mosomes in both the male and the female elements are reduced to half of
the normal number, which is again restored upon fertilization, when the
male and female chromosomes are united. Somewhat later Karl Rabl (1853-
1917), professor at Leipzig, detected the individuality of chromosomes: that
in a cell every chromosome originates from a given chromosome, like itself
in form and size, in the mother cell. And finally, in 1901, the American
W. S. Sutton discovered the so-called accessory chromosome, which at the
nuclear division assumes a place for itself. All these facts have played a de-
cisive part in modern heredity-research and will be further developed later
on in this work.
As a result of these investigations into fertilization,^ very briefly re-
ferred to above, our knowledge of the phenomena of life was so considerably
enhanced that it is difficult to overestimate its value; this not least because
the same fertilization-phenomena were established in the vegetable at the
same time as in the animal kingdom: the union of male and female nucleus,
the reduction of the chromatin, and the individuality of the chromosomes;
all these processes take place with a certain number of modifications, but
on the same principle in every multicellular organism. Life has thereby been
given a uniformity far more demonstrable and real than the hypothetical
common descent of Darwinism. Even in the lowest unicellular organisms,
whether they belong to the animal or the vegetable kingdom, a similarity
in their evolution has been definitely established. We shall now proceed to
discuss these forms.
3 . Microbiology
The Darwinists of the earlier school, chiefly Haeckel, largely interested
themselves, as we have seen, in the very lowest animal forms; it was expected
that they would produce fresh ideas in regard to the origin of life upon the
earth, discoveries that would fill the gap between living and lifeless sub-
stance and would thus make the great evolutional series in the universe
entirely uniform. These expectations, however, whether associated with
Huxley's bathybius slime or with Haeckel's Monera, have not been fulfilled;
bathybius turned out to be a lifeless calcareous deposit, and in the Monera
have been found nuclei and other organic details giving evidence of ordinary
^ In an article entitled " Dokumente xur Geschkhte der Zeugungshhre" (in Archiv fur micro-
scopische Anatomic, Bd. 90), O. Hertwig has given a comprehensive account of the history of
fertilization-research up to the year 1917.
MODERN BIOLOGY 545
cell-Structure. Indeed, the cellular structure of these lowest organisms has
proved to be highly complex, in many of them competing with the funda-
mental elements in the highest organisms. Thus there remains nothing to
be done beyond widening, by strenuous intensive research, our knowledge
of these beings, whose vital manifestations have in many respects proved
of service in answering questions of the greatest theoretical interest, not to
speak of the important practical problems that have been solved through
an extended knowledge of the subject.
Protozoa
Of the unicellular organisms, the Protozoa have, as we know, been longest
known; the earlier progress made in this field of research has been described
in a previous chapter. Of those who have worked at the subject at a later
period the most conspicuous and successful investigators and distinguished
teachers were Biitschli and Richard Hertwig. The Protozoa have proved to
possess a wealth of different forms and structures, both in protoplasm and
nuclei, which has provided the science of general cytology with an invalu-
able material for purposes of comparison. Biitschli's investigations into their
plasma and the changes that take place therein in different stages have al-
ready been mentioned above. Earlier naturalists were inclined to see in the
Protozoa radically undifferentiated plasm, but this assumption has been ut-
terly disproved by experience; on the contrary, the higher unicellular ani-
mals possess a great number of plasmic formations of a markedly organic
character — cilia and flagella, vacuoles and muscular fibrillas. And their
vital manifestations have after careful investigation been proved to exist in
undreamt-of numbers; their irritability and way of reacting to different
impressions offer an important field for experimental biology to investigate.
The nuclei of the Protozoa have been of special interest owing to their
immense wealth of form, to which the higher animal cells have not attained,
and owing to their correspondingly numerous functions. It is in this sphere
that R. Hertwig has made his most important contribution to the advance-
ment of biology. To start with, he has demonstrated that the nucleus in the
Protozoa contains the same constituents as the cell-nucleus in general —
chromatin, linin, and nuclear body. Moreover, the chromatin substance is
in many cases found to be divided up in the cellular plasm in the form of
granules or a network, and sometimes the nucleus is entirely incorporated
in this latter — a phenomenon that is reminiscent of bacterial chromatin's
being invariably distributed over the cellular body, and that at the same
time explains part of Haeckel's Monera. Upon the presence of the nucleus
depends the Protozoa's capacity for assimilation; a fragment of such a cell
without the nucleus would perish for lack of metabolism. Of still greater
importance is the condition of the nucleus in the propagation of Protozoa;
since this generally takes place through division, the process is started by
546 THE HISTORY OF BIOLOGY
the nucleus, which then displays a number of different phenomena in differ-
ent forms: on the one hand, direct division through interlacing; on the other
hand, a regular mitosis, with a division of the centrosome and spindle-
formation; and, between the extremes, a number of transition forms. After
the division the nucleus and plasma grow at different speeds until a certain
ratio of bulk arises between them; then a fresh division takes place. This
"nucleus-plasma relation," as R. Hertwig calls it, is thus a decisive factor
in reproduction and not, as formerly supposed, a growth beyond the nor-
mal standard, for this can vary considerably even in the same species. Still
more remarkable are the phenomena that R. Hertwig discovered upon the
conjugation of the Protozoa — in the fusion of two individuals which pre-
cedes division in certain circumstances. In many Protozoa there exists, besides
the ordinary large nucleus, a small nucleus, called the micronucleus, which,
previous to conjugation, divides itself twice; three of the divided nuclei per-
ish, while the fourth unites with the corresponding nucleus in the conju-
gating neighbouring cell, whereupon out of the unifying product fresh nuclei
are formed in the cells, whose large nuclei meanwhile disintegrate. In the
three disappearing divided nuclei R. Hertwig has seen the equivalent of the
polar bodies in the eggs of higher animals, while the likewise moribund
large nucleus has been held to correspond to the body in a higher animal,
which dies, whereas the sexual cells, here equivalent to the conjugated
small nuclei, reproduce the life-form. Whether or not these comparisons
may perhaps have been carried too far, the future must decide; it is certain
that through them a number of vital phenomena of general interest have
been viewed in an entirely new light, and the uniformity of the fundamental
phenomena of life has received further confirmation.
Bacteriology
Of even greater importance, however, has been the progress made in the
sphere of bacteriology; it was during this period that light was thrown
upon the part played by bacteria as producers of disease and that their bi-
ology was discovered. Theories had long been in circulation that minute
living "seeds of disease" were the causes particularly of the great plagues;
such a hypothesis had been set up during the Renaissance by the Italian
physician Girolamo Fracastoro (1483-1533); Linnsus, it will be remem-
bered, had embraced similar ideas; these theories had been encouraged by
the discovery of yeast-fungi in the eigh teen-thirties; Henle had been spe-
cially interested in "parasites" as producers of disease, and as a proof of his
assumption of such a cause of disease he had formulated the principle : con-
stant existence, isolation from foreign interference, reproduction of the form
of disease by means of the isolated parasite. These conditions, however,
were found to be difficult to fulfil; even Pasteur, who was nevertheless the
founder of modern bacteriology, had not succeeded in finding a means of
MODERNBIOLOGY 5 47
safely isolating the micro-organisms that were to be examined and thus
obtaining pure cultures of them. He had certainly adhered to the idea of
constancy of species in the micro-organisms, but other investigators of the
highest reputation had maintained in contrast thereto the "pleomorphism"
of these beings — that one form could pass unrestrictedly into others of an
entirely different nature; this had been the view of Lister, the famous in-
ventor of the antiseptic bandage, as also of the well-known botanist Nageli,
for reasons which will be mentioned later on. It was in these circumstances
that Koch made his important contribution to the development of bacteri-
ology.
Heinrich Hermann Robert Koch was born in 1843, the son of a miner
in the Harz mountains; he studied at Gottingen under Henle and became
a district doctor in a provincial town in Posen. In that district there was a
serious outbreak of anthrax among the cattle, and the young doctor was
thus faced with the problem of this disease. At an early date a French
physician, Casimir Joseph Davaine (i8ii-8i), a practitioner in Paris, had
discovered small stick-shaped formations in the blood of animals affected
with anthrax and through experiments had found them to be producers of
the disease, but had not succeeded in ascertaining their course of evolution
and method of distribution. Koch took up the problem for fresh treatment;
after victoriously struggling against the difficulties that a provincial doctor
always experiences when he proposes to carry out experimental research-
work, he succeeded in elucidating the entire evolutional history of the
anthrax microbe; how, when introduced into the blood of an animal, it prop-
agates by repeated division on a vast scale, and then, when the animal has
died, these microbes are converted in favourable circumstances into spores
possessing great powers of resistance to external influences and the ability,
after migrating into a fresh animal host, to start the process of evolution
all over again. Koch's genius in these experiments lay in the simple and yet
extremely effective technique that he worked out; indeed, it was in this
sphere that he afterwards won his greatest successes. The anthrax microbe
was at first cultivated in a damp chamber in serum, but Koch soon invented
the method of planting bacteria on a gelatin solution; on this substratum,
which could be made solid or liquid at will by a slight alteration of tem-
perature, it was easy to isolate the bacteria and produce absolutely pure
cultures. The method, which in its simplicity is one of the most brilliant
inventions of modern times, has been the foundation on which the whole of
present-day microbe-research has since then developed. But, in addition to
this, Koch introduced the aniline-dye method into the study of bacteria, a
method which since that time has been perfected in many ways and one where-
by innumerable, otherwise invisible micro-organisms have been discovered
and described; and, furthermore, he invented the microscopical illuminating
548 THE HISTORY OF BIOLOGY
apparatus, constructed to his order by the physicist Abbe, of Jena — nowa-
days an indispensable aid to all who work with strong magnifications.
Koch's first discoveries won him immediate fame; he was elected a mem-
ber of the Kaiserliches Gesundheitsamt in Berlin, whose leading force he
at once became, and had munificent sums placed at his disposal. During the
period that followed, he was responsible for two great achievements — the
discovery of the tubercular bacillus and the cholera microbe. In the case
of the former he tried to produce a specific cure in tuberculin, but, as is well
known, without success. Extremely valuable, on the other hand, was his
reform of the technics of disinfection : the abolition of the earlier ineffective
means of disinfection, such as fumigating with sulphur, and spraying with
carbolic, and the substitution of new experimentally tested and therefore
effective methods.
Koch' s pupils
Koch's activities included also the training of a host of pupils; from all
countries there flocked to his laboratory students, who have since diffused
his methods everywhere. Among these may be named F. J. S. Loffler (born
in 1851), the discoverer of the microbes of diphtheria and swine-fever, and
Emil Behring (1854-1917), professor at Marburg, the founder of serum
therapy. In his later years Koch himself was the accepted authority on
everything concerning problems of infection, and he undertook many voy-
ages, especially to the tropics, with a view to investigate infectious diseases
existing there and to try to find a cure for them. Of a despotic disposition
and spoiled by his early successes, during his last years he did not always
take into account the most recent discoveries, nor who had made them,
which sometimes resulted in disputes that proved of little benefit for the
advancement of science. He laboured, however, up to the last, in spite of
impaired health; he died in 1910 of paralysis of the heart.
While thus the disease-producing micro-organisms were giving rise to
an entirely new branch of research, the yeast-fungi, which were allied to
them, likewise became the object of close study. The pioneer in this field,
next to Pasteur, was the Dane Emil Kristian Hansen (1841-1909). Born of
a working-class family, he became at first a secondary-school teacher, ma-
triculated when he was near the age of thirty, and afterwards applied himself
to the study of chemistry and biology. When the famous Karlsberg laboratory
was instituted by the brewer Jacobsen, of Copenhagen, Hansen became its
leading force, and, in compliance with the wishes of the founder, devoted
himself entirely to the study of the fermenting process. In this sphere he
created what has ever since been the accepted technology of the subject;
in particular, he perfected the pure cultivation of the yeast-fungi by an in-
genious method of isolating a single specimen of these organisms, which
occur in masses and are only visible under the strongest magnifications; by
MODERN BIOLOGY 549
allowing this specimen to reproduce itself there came into being a "pure
line" of yeast-fungi, possessing fully controllable characters. This technique
of yeast-cultivation has entirely reformed the brewing industry and the
manufacture of yeast. Moreover, Hansen has added much of value to our
knowledge of the enzvmes, which play an active part in fermentation phe-
nomena, and a number of other kindred manifestations. A new conception
of the process of fermentation has been produced by Eduard Buchner (1860-
1917), professor of chemistry at Berlin. He has proved that the alcoholic
fermentation of sugar is not, as was hitherto believed, caused directly by
vital action on the part of the yeast-fungi, but that these organisms produce
a chemical ferment which brings about the yeasting and which can be iso-
lated and made to function even in the absence of the fungi. As a result of
this discovery the classical fermentation-theory set up by Pasteur has been
considerably modified and has been transferred from the sphere of biology
to that of chemistry. A number of other phenomena in the same category will
be discussed later on in this work.
Of far later date than the knowledge of bacteria and yeast-fungi is our
knowledge of another group of organisms, which are usually referred to
the animal kingdom and which have been found to resemble bacteria in
being producers of disease — namely, the Sporozoa. Even Meckel was aware
that the well-known disease "ague," or malaria, was accompanied by a
peculiar darkening of the blood corpuscles and of certain other tissue ele-
ments in the infected subjects. But the disease itself was considered to be
"miasmatic" — that is, due to poisonous vapours emanating from the hu-
mid districts in which it occurs. Then the French army surgeon Alphonse
Laveran (1845-19x0, while serving in Algeria in 1880, discovered that the
said pigmentation is caused by a parasite which occurs under various forms,
but which, owing to its mobility, he thought belonged to the animal king-
dom. His accurate description of the newly-discovered producer of disease
was worthy of the closest attention, but it threw no light on the causes
of its distribution. This point was definitely answered by the Englishman
Sir Ronald Ross, who was born in 1857 in India, where he was serving as
an army surgeon, and who was afterwards elected to a professorial chair at an
institute of tropical diseases at Liverpool. He discovered the alternation of
generation in the malarial plasmodium: how, after developing in the human
blood, it is absorbed by blood-sucking mosquitoes, is conveyed from the
mosquito's intestinal canal into its salivary glands, and thence passes into
the blood of human beings, who thus become likewise infected. An impor-
tant contribution to the problem of malaria has been made by the Italian
Giovanni Baptista Grassi (1854-19x5), professor of zoology at Rome, who
made his name especially on account of the effective measures of protection
he adopted against malaria in his own country, which was so terribly
550 THE HISTORY OF BIOLOGY
ravaged by that disease; by excluding the mosquitoes from human dwellings
and eradicating their larvae, he has succeeded in making inhabitable dis-
tricts that were formerly dangerous to live in, and, further, his exhaustive
studies of the biology of the malarial mosquito have made it possible for
other countries also to take energetic measures for its extermination.
A number of other parasites have latterly been discovered and described,
as, for example, the Flagellata, which produce in tropical Africa the fatal
sleeping-sickness, which is transmitted from one person to another by the
tick; and, further, the producer of the cattle-plague, also an African disease,
transmitted by the tsetse-fly, which had made cattle-breeding impossible in
extensive districts. These parasites were specially studied by Koch and his
pupils.
To the beginning of our century belongs the discovery of Spirockate
pallida, the carrier of syphilitic infections, one of the most dangerous ene-
mies of man. It was discovered by Fritz Schaudinn, who has thereby en-
sured for himself a place in the cultural history of the world. Born in 1871
in East Prussia, he studied at Berlin, and after taking his doctor's degree
he was given an appointment at the Gesundheitsamt. Labouring under con-
stant difficulties and in frequent dispute with the old despotic Koch and other
bureaucrats in the Civil Service, who neither would nor could appreciate the
value of his ideas, he worked his way up to a brilliant reputation as a micro-
biologist. It was not until shortly before his death that he received the per-
manent post worthy of him as head of a research institute at Hamburg. He
made valuable contributions to our knowledge of the life of the malarial
parasite; by means of experiments upon himself he studied the dangerous
Amceba histolytica, the producer of a serious form of intestinal catarrh — an
experiment that cost him his life. He also published the valuable results of
his researches into the reproduction of the Foraminifera and Heliozoa. The
above-mentioned discovery of Spirochate pallida he made in the year before
he died. Moreover, he had a number of distinguished pupils, as, for instance,
M. Hartmann, born in 1876, who took up for further study his theoretical
research-work on the Protozoa, and S. Prowazek (1875-1916), who con-
tinued his work on the disease-producing Sporozoa.
4. Vegetable Morphology
Development from romanticism to exact investigation
It is necessary to take a brief glance at the method of morphological re-
search as applied in the sphere of botany, especially in view of the part
played by plants as a basis for modern evolutional theories. For this pur-
pose we must go back to the period before the appearance of Darwinism,
MODERN BIOLOGY 551
when romantic idealism still prevailed in biological research-work. At that
time Goethe's spiral visions and metamorphosis theory were still playing
their part as foundations on which various naturalists based their concep-
tions of the form and growth of plants and the position and development
of their leaves. These imaginings produced appalling confusion in ideas and
theories. "It is remarkable," says Sachs in his Geschkhte der Botanik, "that
as soon as there was any mention of the metamorphosis of plants, even
gifted and clever men gave way to nonsensical gibberish." But even clear-
thinking investigators sought to solve these problems from a purely ideal
point of view; the position of the leaves of the plants was created by an
idea which expressed itself in mathematically formulated relations between
the leaves. Karl Friedrich Schimper (1803-67), for the greater part of his
life a private scholar, was one of these speculative plant-morphologists; he
expressed the "spiral tendency" in the position of the leaves by means of
a serial fraction. His ideas were further developed by Alexander Braun
(1805-77), who studied at Munich, among others under Schelling, and
finally became professor of botany at Berlin and a distinguished teacher.
Among his disciples was Haeckel, who highly admired him and was largely
influenced by him in the romantic direction. Braun, who was otherwise a
specialist of some merit, recorded his morphological speculations in a trea-
tise iJber die Verjiingung in der Nafur, a curious blend of exact knowledge and
romantic imaginative thought. The work contains a number of, for the time,
excellent studies of lower plants, especially unicellular Algas, the growth
and reproduction of which are carefully described. Upon these observations,
as well as some studies of the position of the leaves in buds and flowers and
on the stem of higher plants, is based a "living view of nature," which
tries to find in nature " nicht bloss die Wirkung toter Krafte, sondern den Aus-
druck lebendiger Tat.'' This conception of nature is based on "rejuvenation"
as the driving force in life, whereby the old is constantly being converted
into a new: the child's "old" milk-teeth into new ones, the "old" pupa
of the butterfly larvns into a new butterfly, to say nothing of the spring's
rejuvenation of leaves and herbs, which, of course, gave rise to the whole
of this speculation. "The spirit that develops in man is not outwardly united
to nature, for its appearance is already indicated in the lower stages of nat-
ural life, especially in the animal kingdom; rather the spiritual life is the
purest representation of the same basis of life as that which in previous stages
confronted us as natural life." This, of course, is pure natural philosophy;
it is no wonder, then, that Goethe's metamorphosis doctrine finds its appli-
cation here, both in ascending and descending metamorphosis and in the
spiral arrangement of the leaves, which on the model of Schimper is expressed
in mathematical formulas. It is strange to note how exact observations are
mixed up with this fantastic terminology, especially when it is applied to
55X THE HISTORY OF BIOLOGY
the then newly-discovered cytological details. " Alle Verjungungen im Zellen-
leben sind mit einer mehr oder minder tiej eingreifenden Entbildung der bereits be-
festigten und der Fortbildung understrebenden Telle der Zelle verbunden.'' By this
'^ Entbildung" is meant simply the dissolution of the cell membranes upon
the segmenting of the cells, which, of course, represents the "Verjungungs"
phenomenon. These same principles are also applied to vegetable geography,
and the system becomes merely a link in this magnificent unity whereby
"the whole of nature's course of development from the first manifestations
of life through an infinite number of rejuvenations gradually rises to the
emerging of man." All this speculation is interesting as being an intermedi-
ate link between the old natural philosophy in the spirit of Goethe and
the new philosophy of Haeckel, who differs from his master only in his
materialistic tone, though but little in fact. Haeckel's symmetry ideas in
particular are certainly in imitation of Braun.
Exact research, however, must eventually come into its own even in
vegetable morphology. The scientist who has contributed more than anyone
else towards producing an exact conception of the forms and development
of plant life is Nageli, though even he was in close contact with the old
idealistic philosophy. Carl Wilhelm Nageli was born in 1817 near Zurich,
where his father was a physician, and it was intended that he should be
trained for the same profession at the college in his native town. His at-
tendance at the lectures of the aged Oken, however, induced him to take
up a more speculative career, which he was finally permitted to do. He
studied botany at Geneva under de Candolle and wrote as his dissertation
a work on vegetable classification; he then went to Berlin and spent a couple
of years studying Hegel's philosophy, which, according to his own state-
ment, did not attract him very much, and he finally spent some time working
at Jena under Schleiden. He was a friend of Kolliker and accompanied the
latter on a trip to Italy, afterwards becoming professor, first at Freiburg,
then at Zurich, and ultimately at Munich, where he spent the rest of his
life in work of an unusually many-sided and productive character. Since his
childhood his health had been poor, but he worked with indomitable en-
ergy up to the last ten years of his life, when sickness compelled him to
abandon his activities. He died in 1891. Ill health brought with it an
irritable temper, which made it difficult to associate with him, either as
a teacher or as a man of science; indeed, his personal pupils were few in
number, but the influence exercised by his writings was all the greater. In-
deed, he must without doubt be counted among the foremost botanists of
the century, and that, too, in many different spheres, being at the same time
anatomist and cytologist, morphologist and systematist; moreover, his natu-
ral-philosophical speculations have proved of deep significance.
MODERN BIOLOGY 553
Ndgeli' s cytological investigations
Nageli was certainly greatest as a cytologist; his studies of the dividing
of the pollen-grains and of the unicellular Algas have already been men-
tioned as epoch-making in their sphere, and through them he became one
of the pioneers of modern cytology. We may also include in this category
his studies of the sexual reproduction of the cryptogams, a problem that has
largely been elucidated by him. Nevertheless, his cell research had its weak-
nesses; the fact that he long maintains the old belief in independent cell-
formation is of less importance in this respect than the influence which he
permitted his theoretical speculations to exercise on the observations that
had already been made. In the field of vegetable anatomy a series of essays
that he wrote on the growth of the stem and root forms the basis of our
present-day knowledge of the subject. As a systematist he was especially
occupied in studying genera that possess abundant forms, but are difficult
to elucidate, chiefly Hieracium, the numerous and mutually overlapping mi-
crospecies of which he sought to explain by means of both natural observa-
tions and horticultural experiments. His experiences in studying this difficult
genus led him to speculate upon the term "species,"^ which formed the
basis of his evolutional theories. As a plant-physiologist he distinguished
himself chiefly in his investigations into the growth of starch granules,
whereby he laid the foundation of our knowledge of that curiously organized
structure in these elements of stored nutrition, which, as far as their chemi-
cal composition is concerned, are comparatively simple. Even in this line
of research, however, he became involved in theoretical speculations of that
abstract kind which had interested him since his youth.
In a treatise Uber die Aufgabe der Naturgeschichte, dated 1844, Nageli has
given an account of the theoretical standpoint from which he started. He
lays down as the aims of natural research, firstly the discovery of fresh facts,
and secondly the creation of new laws of thought. His interest in these
latter are clearly reminiscent of his studies in the Hegelian school, referred
to above. True, he indignantly repudiates the accusations of Hegelianism
that were directed against him, but the likeness is nevertheless unmistakable
and gives his speculations a character all its own, which is strongly diver-
gent from, for instance, Haeckel's; while the latter speculates upon forms of
symmetry, the psychic qualities of matter, and other ideas reminiscent of
Schelling, Nageli is ever seeking to create fixed categories of thought, pref-
erably with reference to mathematical deductions. Above all, he strives to
create "absolute ideas," in which the various phenomena are to be defined.
All life is movement, and so all biology must be evolution, and from the
' In contrast to Lamarck, Darwin, and even Haeckel, Nageli speaks in every way depreciat-
ingly of Linnaeus and his work for the advancement of classification. As is well known, these
views, which have but little justification, have since recurred in many German botanists.
554 THE HISTORY OF BIOLOGY
evolution of the individual it is possible to conclude that of the species.
The species is a summary of all similar individuals and, as such, an absolute
idea; similarly, there are many circles, ellipses, and other geometrical figures,
but their ideas are essentially different — and in the same way the species
have no intertransitional forms. "Die absolute Verschiedenbeit der Arten scheint
mir durch die Erjahrung hinldngliclj bestdtigt, und allgemein genug angenommen,
um auch ihrerseits die Absolut he it der Begrijfe ■zii bestdtigeti." And like the species
the higher systematical categories also possess their absoluteness; the vege-
table and the animal kingdoms have no transitions, for " dieser Annahme ivider-
spricht schon die Absolutheit der Begriffe. ' ' If this is not Hegelianism, it is at any
rate very near it.
His micella theory
The ideas expounded in this work of his early years proved in many respects
to have a decisive influence on Nageli's future development. In his work on
starch granules he presents his once famous micella theory, which again
shows his tendency to transfer the deductions of geometry to biology. Ac-
cording to this theory the cells and their derivatives are composed of parti-
cles called "micellas," which are supposed to be composed of a number of
molecules, possess a regular crystalline form, and in a dry state keep close
to one another, owing to their mutual attraction; in certain circumstances,
however, the micellas attract water, which penetrates in between them and
surrounds each one of them, and through this action arises tissue. This the-
ory, which was irreconcilable with the findings of physics and consequently
had to be abandoned even during the lifetime of its originator, shows how
he strove to compare animate and inanimate matter; in actual fact he denies
any principial difference between them. In consequence he believed also in
spontaneous generation, stubbornly maintaining that doctrine throughout
his life; he certainly recognized Pasteur's experiments, but he considered that
they did not demonstrate the impossibility of spontaneous generation, and
to his mind spontaneous generation is ' ' nicht eine Frage der Erjahrung und des
Experiments, sondern eine aus dem GesetT^e der Erhaltung von Kraft und Stoff fol-
gende Tatsache." In his youth he believed in the spontaneous generation of
unicellular sponges and believed that it could be demonstrated by experi-
ment; when he did not succeed, he assumed the spontaneous generation of
very primitive unicellular creatures, and in his old age he continued his re-
treat, inasmuch as he assumed as products of spontaneous generation a kind
of extremely primitive life-units, whereof countless numbers go to form one
cell, and which he termed " probia." But he was undeniably more consistent
than Haeckel in not moving spontaneous generation back to the beginning
of the world, holding that it could just as well take place now as then,
"for the difficulty of letting a cell arise out of formless chemical substance
is not a jot greater for primeval times than for modern times." We must
MODERN BIOLOGY 555
also take in conjunction with this theory his positive assertion, previously
referred to, as to the pleomorphism of micro-organisms; it cannot, of course,
be denied that the absence of any species-characters in them was a feature
of primitive organization, which would make them closely akin to lifeless
matter.
His descent theory
Nageli's descent theory is also in line with his theory of spontaneous genera-
tion; since spontaneous generation goes on incessantly, it is possible to sup-
pose that the most highly developed organisms are really the oldest, while
the primitive organisms have been evolved later. His descent theory has thus
acquired a decidedly polyphyletic character, and, strictly speaking, it does
not presuppose any transition from one species to another. The most inter-
esting feature in Nageli's phylogenetical speculation — recorded in an essay
on the genesis of the natural-historical species (1865) and in a large work,
Mecbanisch-pbysiologische Theorie der Abstamtnungslehre (1884) — is his criti-
cism of Darwin's theory of selection. In the course of his experiments on
the Hieracium he had discovered that the external conditions of life which
cause the struggle for existence do not alter the life-types; species that are
placed in new conditions of life do not assume any similarity to kindred
species previously brought under these conditions. Natural selection, there-
fore, cannot possess any form-building power; it does exist, but has only an
extenuative effect on middle forms. Instead, evolution takes place out of the
inner being of the organisms in virtue of an internal force, which Nageli
most frequently terms " Vervollkommmtgskraft," and once, in imitation of
Blumenbach, " Nisus formativus," a force by means of which the development
is led in a certain direction, not, as Darwin holds, to variations in every
possible direction. This force, however, is by no means a special life-force;
on the contrary, it is compared with the inertia in inorganic nature; just
as a rolling globe goes on until it meets an obstacle, so organic evolution —
it, too, being a movement — advances until an obstacle comes in the way.
These obstacles can be either the struggle for existence, or else direct ma-
terial influence due to irritation; according to Nageli the ruminants have
got horns as a result of striking their foreheads against one another — an
explanation in the spirit of Lamarck.
His heredity theory
In connexion with his doctrine of descent Nageli propounds a heredity the-
ory of his own. At variance with Haeckel's view on the undifferentiated
character of the egg-cell he definitely maintains that the egg-cell is as com-
plicated as the creature which is to be evolved therefrom; the qualities of
the coming individual are all united in the egg-cell. But because this latter
and the sperm cell, in spite of their difference in size, have an equally large
share in the qualities of the new individual, these qualities cannot be
556 THE HISTORY OF BIOLOGY
allocated to the whole protoplasm of the egg; there must be a certain part
of them that is particularly responsible for the specific qualities. This con-
stituent of the cell Nageli calls idioplasma; he believes that through segmen-
tation it is imparted to every fresh cell and gives the latter its character;
it is through it that every organism is such as it is and not otherwise. The
idioplasma is, according to Nageli, a solid body, not semi-fluid like the rest
of the cellular mass, and it has, of course, its peculiar composition of micella,
the shape and size of which give rise to the most subtle calculations. All
evolution consists in changes in the micellas of the idioplasma, and these
changes go on incessantly, although they are not at once perceptible, for
the energy amassed through these changes is released intermittently, and
therefore the alterations in species likewise take place, not gradually, but
suddenly.
From the structure of the idioplasma Nageli gradually passes to atomic
structure in general, and he here becomes involved in speculations as to the
atoms' being composed of still smaller particles, which are called " amera";
of these latter the simple chemical basic elements are composed, and Nageli
builds up a kind of phylogeny for these elements, according to which the
heavy metals must have originated first, and afterwards the other elements
in succession. Further, he speculates upon the form of the atoms, upon ethe-
real atoms, ethereal heat, upon the impossibility of entropy, and various
similar subjects, which contemporary physics and chemistry had naturally
passed over in silence.
His influence
Nageli's mechanical-physiological theory was his last work, so that he
concluded his life's activities, in spite of his expressed intention to deal with
natural phenomena on a mathematically exact basis, in a mass of thought-
constructions of just as impractical a nature as those of the master of his
youth, Hegel. His influence, however, has been of deep significance, not only
on account of the immense number of important facts that he established,
but also in the purely theoretical sphere. He was the first unreservedly to
venture to reject the doctrine of natural selection as the sole cause of the
evolution of life and to demand that it be replaced by another theory capable
of producing a more convincing confirmation by way of observation and
experiment. The " Vervollkommungskraff" on which he would base his ex-
planation of the origin of species was really nothing but a word, but behind
it there lay at any rate an insight into the fact that evolution is a quality
in life itself, not a movement that is thrust upon living creatures from out-
side. And in connexion therewith Nageli points out that life need not
necessarily evolve as a result of minute imperceptible variations, but the
changes might just as well take place suddenly and on a larger scale — an
idea which is certainly not very strongly brought out in him, but which
MODERN BIOLOGY 557
nevertheless afterwards survived; de Vries especially inherited it through
his theory of mutation, but, above all, Nageli's idioplasma theory was an
idea that was utilized by subsequent investigators with much profit. Here,
again, he really only invented a word, but the idea of a special substance's
being the bearer of the cell's hereditary qualities received remarkable con-
firmation in the above-described discovery of the role played by chromatin
in cell-division and its importance for the vital processes of the cell in gen-
eral. What prevented Nageli himself from drawing from his speculation
conclusions of practical value was undoubtedly his belief in ' ' absolute ideas ' '
and the derivation of facts from them — a belief that he never really suc-
ceeded in eradicating. Herein, too, we must obviously seek the cause of his
attitude towards Mendel, violently criticized at a later date. The latter had
reported to him the results of his epoch-making experiments and received
in reply an inquiry as to whether the formulas he had set up were not "em-
pirical rather than rational." In these words is clearly shown the weakness
of Nageli's abstract-speculative method: he could not grasp Mendel's incon-
trovertible results based on fact, since they did not agree with his own the-
ories, and the correspondence, which went on for some time, though in
courteous terms, produced no result. With all his weaknesses Nageli never-
theless stands out as one of the foremost biologists of his time, and his
ideas had an influence long after his death.
Among Nageli's pupils the first that deserves mention is his fellow-
countryman, Simon Schwendener (1819-1919), for a long time an assistant
to his master and finally professor at Berlin. Of his works should be men-
tioned one entitled Dasmechanische Prin^ip im anatomischen Bau der M.onokotylen.
In this he describes the mechanical functions of the cells and tissue elements
and shows how the structure of the plant closely follows the general laws
of mechanics governing its sustaining power and strength. In doing so, how-
ever, he sometimes interprets the structure and functions of plants from
a too narrowly mechanical point of view. Thus, for instance, he sets up a
mechanical theory in regard to the position of the leaves, wherein he exam-
ines the above-mentioned idealistic spiral theory and finds that the leaves'
spiral position is caused by conditions of mechanical stress and is altered
if the stress alters. In spite of its one-sidedness, however, this work contrib-
uted in its own sphere towards overcoming the romantic belief in an idea's
being the cause of a natural phenomenon and substituting a mechanical ex-
planation. Schwendener's works in the sphere of lichenology, however,
caused a still greater sensation than the above investigation. Hitherto the
lichens had formed a class in the vegetable kingdom by the side of Algas
and Fungi. Schwendener now declared, as the result of a series of micro-
scopical investigations, that the lichens are really a kind of double organ-
isms, consisting of fungous hyphas, in which cells of Algas lie embedded
558 THE HISTORY OF BIOLOGY
and which jointly contribute to the existence of the whole, the Fungi by
forming the substratum, the Algx by assimilating carbonic acid with their
chlorophyll. This discovery, which upon publication aroused the keen op-
position of the lichen-systematists, has gradually received confirmation and
is now universally accepted as correct.
Among those who, as far as the vegetable kingdom is concerned, paved
the way for a uniform conception of its vital manifestations, must also be
mentioned Wilhelm Hofmeister (182.4-77). Born in Leipzig, he was edu-
cated with a view to taking up a commercial career and became a music-
seller in his native town, but he spent his spare time studying botany, and
eventually became a professor, first at Heidelberg and then at Tubingen. His
great achievement is his comparative investigations into the reproduction
of plants, which he carried out while he was still a music-dealer and which
resulted in his being appointed professor. He closely studied the phanero-
gams, as well as vascular cryptogams and mosses, especially observing their
formation, development, and combination of the sexual products, and he
established in all these phenomena a bond of agreement that made possible
in all essential respects the adoption of a uniform conception of sexual re-
production throughout the vegetable kingdom — an achievement that is
all the more remarkable, seeing that his knowledge of the cell was not in
advance of the stage at which his own period had arrived. Hofmeister's
work on the reproduction of plants was followed up by several later natu-
ralists. Among these may be mentioned Nathanael Pringsheim (1813-94),
at one time professor at Jena and then a private scholar in Berlin. He found
out the method of reproduction of the Algas and published several valuable
works on plant physiology. Also, Heinrich Anton de Bary (1831-80),
professor at Strassburg, who discovered the sexual reproduction of the Fungi
and the alternation of generation in the rust fungi, and also solved a
large number of important problems in the sphere of mycology and
bacteriology.
Considerations of space forbid our continuing the account of the develop-
ment of plant morphology up to modern times; in fact, all the details will
be found in the text-books on the subject. We shall therefore proceed to
another branch of biological research, which has also played an important
part in modern times.
5. Geographical Biology
In the foregoing, Humboldt and Wallace have been named as founders, in the
modern sense, of vegetable and animal geography. Like all other branches
of biology, these fields of research have in our day become highly specialized.
MODERN BIOLOGY 559
while fresh fields have been opened up for the employment of their methods
on an extensive scale. Among these novel spheres we note not only new
land-areas — it was not until our own period that the entire globe can
be said to have been explored and described — but also to a still greater
degree the oceans, the deeper areas of which have only recently been known
as regards their physical character and conditions of life. Our knowledge of
them has been gained partly through the work carried out at the zoological
marine stations, of which the most famous in recent times has been that
founded by Anton Dohrn (i 840-1 909) at Naples, and partly as the result
of oceanographic expeditions specially equipped for the purpose, among
which may be recalled in particular the important English Challenger expedi-
tion (i 872.-6) and the German Valdivia expedition (1898-9), not to mention
the results obtained by polar expeditions equipped by a number of countries,
both large and small. It is through these voyages of exploration that the
life-forms of the ocean first became known — the life in the vast depths,
whose denizens live in constant darkness and under high pressure and of-
ten assume amazing forms; the actual inhabitants of the vast ocean-expanses,
the so-called plankton fauna and flora, with their often transparent and frag-
ile forms, which are constantly swimming in the water, forms of widely
differing systematic groups; and, lastly, the life that moves around the coasts.
Innumerable workers have devoted themselves to this branch of study, which
has often been carried out under the leadership of committees, national and
international, with the consequence that the work and its results have to
a certain extent acquired an impersonal character. The pioneer in this field
is Karl August Mobius (1815-1908), professor first at Kiel and then in
Berlin. By his great work Die Fauna der Kieler Bticht (1865) he has created
the modern system and methodics of oecology. By way of introduction he
describes the topography of the estuary that he investigated, various sec-
tions of it being surveyed and characterized in regard to position, depth,
and their plant and animal life. He then presents in systematical order the
creatures that inhabit each locality. Others have continued along the path
thus beaten by Mobius. Among them may be named Victor Hensen (1835-
1914), who was professor of physiology at Kiel and in that capacity studied
the structure of the auditory organ, but afterwards he devoted himself en-
tirely to marine research, chiefly with the idea of improving the fishing
industry. He made a special study of plankton life, with particular refer-
ence to its microscopical forms. In order to advance the study of these crea-
tures, which are of importance as food for fish, he worked out a statistical
method of his own. Of others who laboured in this sphere, to some extent
practically important, may be cited the Dane C. G. J. Petersen (born in
i860), who investigated animal life in the sounds and bays of Denmark on
a method of his own, and J. Schmidt (born in 1877), who after lengthy and
560 THE HISTORY OF BIOLOGY
difficult exploration succeeded in elucidating the reproductive process of the
eel — a problem that many before him had tried to solve in vain.
Touching the continental life -forms, the classification into large geo-
graphical regions already drawn up has on the whole been retained, and
both plant and animal geographists have for the most part devoted them-
selves to the study of conditions within smaller areas belonging to these re-
gions. Among investigators of this category in the sphere of animal life may
be mentioned the explorer Karl Semper (1831-93), professor at Wiirzburg,
who studied the problem of the life-conditions of animals from various points
of view.
Plant geography: its floristk and fnorphological courses
In the field of plant geography, research has taken especially two courses,
a systematical, which is ultimately based on Linnxus's observations and
theories in connexion with the distribution of the plant species, and a mor-
phological, which has its origin in Humboldt's theories on the morpho-
logical association of different vegetable types with different countries and
forms of landscape. These two tendencies have exerted a mutual influence
and have, each in its own way, been influenced by the doctrine of descent
and its attempt to explain the origin of species out of conditions of geo-
graphical distribution. And at the same time valuable results were gained
by the comparison between the distribution of existent plant-forms and that
of the corresponding genera and species of earlier geological periods. All
representatives of modern plant-geography have been compelled more or less
to take these conditions into consideration. It is still possible, however, to
trace two main tendencies in this sphere, which nevertheless incessantly
touch and cross one another. The first of these, the systematic or floristic,
which rests upon the systematic entities, treats of the distribution of the
species within larger or smaller areas and their variations in different parts
of one area under the influence of certain factors. It endeavours to find out
the causes of the changes in the character of species in certain localities and
countries and for this purpose studies the migrations of species, such as oc-
cur through the distribution of land and sea in recent times and through
the changes that have taken place in the distribution in earlier ages, in so
far as it has been possible to trace these shiftings of the world's surface.
Further, it examines the distribution of extinct and fossilized species, from
which those of our own time may possibly have originated. Special interest
has been devoted to the immigration of plants in those parts of Europe that
were once visited by the glacial period, as well as those vegetable remains
in the mountain ranges of the polar regions that give evidence of a previous
warmer climate there.
^ Morphological or oecological plant-geography does not investigate the
nature of the flora, but of the vegetation. It works, not with species, but
MODERN BIOLOGY 561
with plant communities, by which is meant plants of very different systemati-
cal categories that, on account of a uniformity in the alimental conditions,
have adapted themselves to living together within a certain area. The aim
of this tendency is to analyse such plant-associations and to ascertain their
relations to the climate, the soil, and other environmental conditions. Both
these tendencies have made considerable progress up to recent times and can
claim a number of distinguished representatives, of whom it is possible only
to name a few. The two previously mentioned English botanists Brown and
Hooker made valuable observations as to the distribution of plants, espe-
cially in extra-European countries. The Swiss Oswald Heer (1809-83) made
a special study of the conditions of the flora of the glacial period and also
of earlier geological strata. There followed in his tracks the Swede Alfred
Nathorst (i850-i91x), who did very creditable work in investigating the
fossil vegetable world of the polar countries. Adolf Engler (born in
1844), professor at Berlin and founder of an important school of plant
geography, has endeavoured, by studying the recent and fossil vegetable
world, to gain some insight into the evolution and changes of the flora, espe-
cially in the temperate countries. August Heinrich Grisebach (1814-79),
professor at Gottingen, sought to carry out an investigation into the influ-
ence of climate on the vegetation and a classification, on a climatic basis,
of the flora in certain areas. The Dane Eugen Warming (1841-19x4) per-
formed a considerable service to science by his study of plant-associations,
which he classified and analysed in respect of plant forms and conditions
of life. By this work he made a contribution to oecological plant-geography
of fundamental importance. Andreas Franz Wilhelm Schimper(i856-i9oi),
professor at Basel, made long voyages for the purpose of studying tropical
vegetation, and from climatological and modern physiological points of view
he worked out the vegetation of the entire globe in his Vjianxengeogra-phie
auf fhysiologischer Grundlage (1898). His geographical and oecological classi-
fications have exerted great influence upon subsequent development.
CHAPTER XVI
NEO-D ARWINISM AND N E O - L A M A R C K I SM
Decline of Darwinism
TOWARDS THE CLOSE of the nineteenth century the influence of Darwin-
ism began noticeably to wane. The signs of this are many: partly
internal, in that the actual theory, as had so often happened before
and indeed always will happen with dominating views, becomes split up into
a number of mutually conflicting tendencies in different directions, and partly
external, in phenomena manifested in the general cultural situation. The op-
timistic belief in progress as a law governing nature and human life, which
prevailed in the middle of the century and formed the basis of the success
of Darwinism, had some decades after been essentially disturbed. The un-
limited progress that was to follow upon political and economic freedom
had proved to be somewhat relative; democracy, which had been introduced
in many countries, had led to disappointments, out of which much capital
had been made by its political opponents, while free competition had called
forth, not a friendly and stimulating rivalry with a universally acknowledged
precedence for the best, but an inimical and severe struggle between rival en-
terprises, social classes, and nations, wherein people sought rather to do one
another the greatest possible injury. It was quite natural that the confidence
in liberalism that had but recently been so strong should in such circum-
stances begin to waver; the belief that progress goes on by itself began to
be regarded as a matter of course; instead men of courage were required to
remove the increasing difficulties. So there arose a long line of opponents
to liberalism, from the strange romanticist Carlyle, with his demand for
hero-worship, to Nietzsche, with his paradoxical "superman" ideal; both
deserve mention as men who made violent attacks on Darwin and his theory.
Their successes in the sphere of literature may thus be registered as defeats
for Darwinism, and they were by no means the only ones of their kind; on
the contrary, there appeared in th nineties a literary tendency that was
wholly intended to be a contrast to the naturalistic literature of the pre-
ceding decade based on natural science. And while the popularity of Dar-
winism among the general public thus began to wane, its champions among
the scientists had to defend themselves against the obstacles that the re-
sults of fresh research placed in the way of the old theory.
As we know, both Darwin and Haeckel had based their doctrines of
561
MENDEL
From Grtgor Johaim Moidel : Leben, Werk und Wirkmig by Dr. Hugo litis,
published by Verlag von Julius Springer, Berlin
RICHARD OWEN
MODERN BIOLOGY 563
descent partly on the theory of variability and natural selection brought
about by the struggle for existence among the variations, and partly on the
assumption of the direct influence of environment upon the individual, and
the inheritance of the changes thus brought about — that is, a Lamarckian
conception. Here at once, in this double explanation, lay the seeds of dis-
sension: one could with prejudice emphasize the idea of selection, or with
equal prejudice maintain the influence of environment. And this was ex-
actly what happened; the period immediately preceding and around the turn
of the century witnessed the birth of the two evolutional schools of thought
called neo-Darwinism and neo-Lamarckism, whose advocates sought to con-
vince the biologists of the absolute validity of their own views. Out of these
two main directions there further originated a number of special attempts to
explain the causes of evolution, so that the situation in which the doctrine
of descent eventually found itself was somewhat chaotic. We shall here de-
scribe some features of this internal dissolution of Darwinism.
In Germany the theory of selection found a highly gifted and power-
ful advocate in the person of August Weismann (1834-1914). He studied
medicine, being a pupil of Leuckart, who inspired him with an interest for
biology. After working for some years as a practitioner he was invited, on
account of a useful treatise on the evolution of flies, which he had written
in the mean while, to be professor at Freiburg, where he laboured until his
death. His special subject was the evolution of the lower animals; in this
field he particularly distinguished himself in his studies of the reproduction
of the Daphniidas, as a result of which he elucidated the peculiar egg-de-
velopment in these crustaceans and the no less curious "cyclic reproduction"
that characterizes them. An ophthalmic disease, however, soon precluded
him from using the microscope and compelled him to apply himself partly
to experimental and partly to purely speculative activities. One result of
this was his strange theory of evolution, which placed him among the very
foremost of Darwin's successors.
Weismann's theory of descent and heredity is based, firstly, on his
above-mentioned special investigations, and secondly on Nageli's idioplasma
theory, referred to above. Nageli had sought for a material substructure for
the inherited dispositions, out of which are developed in every individual
certain given qualities, and he believed he had discovered it in his hypothe-
sis of the idioplasma, which, existing equally in the egg and in the sperm,
through their union forms in the new individual the basic material for its
special qualities. Weismann, who as a result of his studies and his own re-
search work had acquired a deep insight into contemporary cytological
knowledge, came for that very reason, when he was forced to devote him-
self to purely theoretical speculations, to take up the question of cell-struc-
ture as a basis for the evolutional theory that Darwin and his school had
5 64 THE HISTORY OF BIOLOGY
promulgated. In a series of lectures and monographs dating from the eighties
he endeavours to find out what it is that produces heredity in a biological
sense; how, he asks, are we to account for the characteristic in organisms
of transmitting to their offspring their own essential being, for the fact that
from the eagle's egg is invariably hatched an eagle, and, moreover, one of
the same species as its parents? In imitation of Haeckel he starts from the
unicellular animals and finds that in these the mutual resemblance of the
different generations is due to the individuals' propagating by division; to
the fact that every infusorian is a segment of a previous one, that there thus
exists in them a " Continuitat des Individuums ." And the same is true of the
multicellular animals in virtue of sexual reproduction; for the individual's
life the sexual cells are without significance, but they preserve the continuity
of the species through the ages; out of them arises in certain given circum-
stances a new individual of the same kind as the old.
Weismann s germinal--plasf7t theory
From this it may be concluded that there exists a special "germinal plasm"
which corresponds to the individual series in unicellular animals and which,
like them, preserves the species by repeated dividing, whereas the corporeal
plasm of the individual gradually falls into decay. Originally the differen-
tiation of sexual and corporeal cells had been due to a division of labour
in the simplest cell-colonies, such as we still see in the primitive colony-
forming animals; for the sexual cells that perform the function of reproduc-
tion contain both germinal and corporeal plasm, which separate when, in
the earliest embryonic stage, the rudimentary cells of the sexual organs sepa-
rate from the rest of the cells. Out of the germinal plasm, therefore, arises
the long series of analogous individuals, and these resemble one another for
the very reason that their form is governed by the character of the germinal
plasm, which is determined once and for all; if changes appear in the exter-
nal bodily form, they correspond to and are induced by changes in the ger-
minal plasm. These changes are brought about by fertilization, in which
the germinal plasm of two different individuals is united; through this "am-
phimixis," as Weismann calls it, is formed a new germinal plasm, with both
the parents' qualities, which accordingly appear also in the offspring. But
if the qualities of the individual are thus due entirely to the germinal plasm,
there can be no possibility of influencing the individual series from out-
side; the organs of the individual that are formed of corporeal plasm can
be influenced by practice, in so far as the germinal plasm has created possibili-
ties therefor, but changes of this kind exercise no influence upon the germinal
plasm. Consequently, Lamarck's theory that the character of the species is
created by the habits of the individual is untenable.
This denial of the heredity of acquired characters became one of the
corner-stones of Weismann's biological theory and he sought in many and
MODERNBIOLOGY 5 65
various ways to procure proofs for his argument. He bred large quantities
of rats, whose tails he cut off at birth, but he never succeeded in finding a
rat born tailless, nor did other malformations brought about by outward
interference ever reproduce themselves. He therefore felt fully justified in
maintaining his standpoint that all changes in the outward appearance of
the individual compared with other individuals are due to changes in the
germinal plasm; that every so-called acquired character is really produced
by a change in the germinal plasm, whereby the body becomes capable of
adapting itself to the different external conditions of life. But how, then,
have the various life-forms arisen in the course of ages? By means of natural
selection, answers Weismann, and by that means alone. The variations that
are brought about especially through the amphimixis of sexual reproduction,
but also through other changes in the germinal plasm, are advanced or re-
tarded by natural selection and thus give rise to new forms, whose germinal
plasm is better adapted to the conditions of existence than that of the old
forms. Natural selection is thus the cause of the evolution of animate beings,
Weismann rejecting Nageli's assumption of internal causes of evolution in-
herent in the organisms themselves, for such a theory "cannot explain the
finality of the organisms. And yet this is the very riddle that the organic
world gives us to solve." Numberless instances are quoted of this adaptabil-
ity, this connexion between form and function, and every instance is likewise
made to serve as evidence of the creative power of natural selection.
' ' The continuity of the germinal plasm ' ' and ' ' the omnipotence of natu-
ral selection" are two phrases in which Weismann's theory of life used to
be summed up. As a result of the former of these ideas — that of the germinal
plasm as the preserver of heredity — Weismann has reached by way of specu-
lation conclusions to a certain extent foreshadowing those that modern
heredity-research has since arrived at by means of exact observation. His
subsequent attempts to expand this theory gave him similarly happy in-
spirations, as when he localizes the germinal plasm — that is, the preserver
of heredity — in the chromosomes of the sexual cells. "The idea in itself
was sound," says Johannsen in this connexion. Even Weismann, however,
succumbed to the danger of basing his conception of a phenomenon on mere
speculation; in his continued efforts to extend his germinal-plasm theory
downwards he works out a highly complicated plan to show the structural
nature of living substance; every one of its minutest entities consists of a
mass of chemical molecules; they are termed "biophores," and he assures
us that they are not hypothetical: "They must exist, for the phenomena of
life must be bound to an entity of matter." Of biophores are composed the
determinants: those units in the germinal plasm that govern the various
qualities in the smallest parts of the individual; the determinants in their
turn build up the ids, which form larger groups of qualities, and these again
566 THE HISTORY OF BIOLOGY
the chromosomes, which unite in themselves all hereditary qualities. With
these we have at last arrived at something that can be observed; in fact, the
foregoing has been pure imagination, of the kind that the biologists of
the past century had such difficulty in avoiding when they had to explain
the phenomena of life. Darwin and Haeckel were content each with one
hypothetical unit of life; it can hardly be said that Weismann would have
done any special service to biology by burdening it with three of them.
Theory of germinal selection
However, the germinal-plasm theory and its conclusions, both the ingen-
ious and the false, only served Weismann as a means for proving the doc-
trine that gradually came to mean for him the very corner-stone of biology:
the doctrine of the omnipotence of natural selection. The championship of
this theory, and the fight against that of the inheritance of acquired char-
acters eventually became his chief aim in life; all that could serve his pur-
pose he took to be good, while all that militated against it was rejected.
He went through many a hard struggle on behalf of his favourite theory;
in the nineties he was especially attacked by Herbert Spencer, who main-
tained the doctrine of the transmission of acquired characters, chiefly for
social reasons; it was, in fact, the precondition of human progress. But from
many other quarters also there arose the cry of "the impotence of natural
selection," and this cry was again taken up after the turn of the century.
Weismann's defence was often somewhat laboured; against Spencer he de-
fended himself mostly on the old argument about the intelligence of the
workers among the bees, which cannot be transmitted by inheritance, since
they are sterile, and which therefore cannot be directly "acquired" either.
It was more difficult to answer the question as to how that finality arose
that shows itself in occasional encroachments upon an organism; how, for
instance, a fracture heals in certain definite ways; the fracture certainly can-
not be traced to natural selection. Here Weismann found support in a theory
that was produced by the afterwards famous experimental biologist Roux,
who in his youth published a work entitled Der Kampfder Teile im Organismus.
Here an attempt is made to explain what Roux calls ' ' functional adaptation
within the organism: that every organ, even every cell, possesses its given
structure, which changes if the conditions of the organ's function are changed,
so that in normal circumstances the life of the body runs its even course;
if this is disturbed by interference from outside, cells and tissues adapt them-
selves as required to repair the damage. This fact Roux considers to be due
to a "struggle for existence" between the cells in the body and even between
the molecules in every cell, each of which strives to force its way forward
at the expense of its neighbours, an effort that is controlled by the general
requirements of the body, the weakest elements being thrust aside and de-
stroyed. This theory, to which we shall revert in another connexion, at
MODERNBIOLOGY 5 67
once won Weismann's keen approval, but it met with opposition from other
quarters; as, for instance, from O. Hertwig, who held that upon the first
division of the egg it should be possible to see something of this struggle
between the cells, but that, on the contrary, the first easily observable em-
bryonic cells show no inclination whatever for mutual strife, but rather
each one has its carefully defined form and place. Weismann, however,
adopted the idea of natural selection within the organism and combined it
with his germinal-plasm theory. In a work entitled Uber Germinalselektion
he declares that between the various parts of the body and their "deter-
minants" in the germinal plasm there exists reciprocal action; if, now,
an organ is not used, its determinants are weakened and annihilated by
the struggle within the organism, and the organ disappears in succeeding
generations; in this way, for instance, the posterior extremities of the whale
have been lost. But, all the same, Weismann comes in this way, although
indirectly, to accept the inheritance of acquired characters, which indicates
that the theory of selection finds it difficult to do without this auxiliary
hypothesis. We shall here leave the omnipotence of natural selection and
pass on to its diametrical opposite, neo-Lamarckism.
Lamarck's theory of the direct influence of habits of life upon the bodily
structure of the individual and its offspring gained strong support towards
the close of the century, especially in France. When the supporters of Cuvier
finally left the arena, it was to Lamarck that people turned for a basis for
their biological ideas. When the belief in the constancy of species had to
give way to the theory of evolution, the form that this was to take was
readily sought from a fellow-countryman, and, moreover, an older man than
Darwin; thus "transformism," as it was here called, could also claim to be
an originally French science. Lamarck's theory found an eloquent supporter
in Alfred Giard. Born in 1846, he studied at the Ecole Normale in Paris
and eventually became professor of zoology at the Sorbonne and head of the
marine laboratory at Wimereux, near Boulogne; he held that post with suc-
cess until his death, in 1908, being especially known for his profound studies
of a number of marine animal forms. Under the characteristic title Contro-
verses transjormistes he collected some years before his death a series of con-
tributions to the problem of descent, in which he examined and further
developed Lamarck's doctrines. According to Giard, evolution proceeds un-
der the influence of two categories of factors; namely, the primary, which
directly influence the individual and indirectly its offspring, and among
which are mentioned light, temperature, food, and relations to other beings
— that is to say, the struggle for existence — and the secondary, which
include everything that is adapted to remove less suitable forms of life —
that is to say, natural selection. Giard now takes upon himself to prove the
existence of the primary factors, and he adduces quite a number of proofs.
568 THE HISTORY OF BIOLOGY
Among the positive proofs he includes a number of experiments carried out
by the physiologist Brown-Sequard, who believed that by interfering with
the nervous system in guinea-pigs he had induced epilepsy in their young;
these experiments, however, have been found by other investigators either
to have miscarried or to have been misinterpreted. Giard has better success
when, in a controversy with Weismann, he declares that the secondary fac-
tors alone cannot explain the origin of the forms of life; he points out a
number of phenomena that cannot be explained by selection alone. On the
other hand, he does not deny the existence of selection, as already mentioned
above; he only considers its importance to be "secondary."
0. Hertwig against natural selection
A FAR more severe judgment than that of Giard and several other Lamarckists
— for example, the famous American paleontologist E. D. Cope (1840-
57) — is passed upon the theory of selection by Oscar Hertwig, who de-
voted the latter part of his life particularly to attacking the common belief
in it. In fact, in his great work of 1916, referred to above. Das Werden der
Organismen, he finally settles his account with that theory and at the same
time gives a summary of the natural philosophy which he had produced in
the course of a long life that had been unusually rich in experience. Being
mainly a cytologist, Hertwig attaches decisive importance to the cell and
its structure as the groundwork for all speculation upon evolution. To him
the cell is the elementary organism and he vehemently sets his face against
all theories of biophores, plastidules, and such lower vital entities. Every
form of life has its peculiar cell-structure : there are in nature as many ' ' species
cells" as species; it is the character of the species cell that causes every form
of life to be what it is and produces descendants of the same kind. Evolution
is regulated in each separate case by the character of the species cell, and
those phenomena that coincide therewith are described as the " biogenetical
law of cause," which precludes the possibility of any such biogenetical prin-
ciple as that which Haeckel conceived; in its embryonic development a mam-
mal certainly does not pass through a series of stages identical with the
lower animals, but rather the egg of every mammal species is just as fully
specialized as the animal itself, and similarly with the embryonic stages. The
tgg contains within itself all the characters of the organism as rudiments;
Hertwig, following Nageli, terms the bearer of the rudiment within the
cell "idioplasma," but he is generally content to speak of the rudiments of
the species cell. Towards the question of heredity he adopts a decidedly
morphological attitude and insists that the material basis of the relative
phenomena must be observed and explored, thereby opposing Johannsen's
physiological view of the phenomena of heredity. He most emphatically
maintains the heredity of acquired characters — that is to say, the metab-
olistic influence of environment upon the hereditary dispositions. In support
MODERNBIOLOGY 5 69
of this assertion he cites a number of experiments, which, however,
have been interpreted differently by modern heredity-research, such as, for
instance, Kammerer's experiments with colour changes in the salamander,
and Tower's experiments in connexion with the evolution of the beetle. If
Hertwig's views on this point approach those of Lamarck, he refuses all
the more definitely to have anything to do with Darwin's theory of selection.
He brings out the latter's weaknesses in a strong light; he especially points
out how much of the theory is borrowed from human conditions and is,
moreover, utterly misinterpreted. The breeder who selects suitable variations
creates nothing new thereby, but only chooses what suits him, and this pro-
cedure has no counterpart in nature — the struggle for existence does not
destroy creatures; the masses that die do so from quite different causes. To
declare that selection favours certain variations postulates mere chance as
an operative cause, but chance is no natural-scientific explanation. And he
views with equal disfavour the above-mentioned theory of the struggle of
the parts within the organism; this, too, is found to rest upon utterly false
conclusions.
Once again Oscar Hertwig attacks the theory of the struggle for exist-
ence and natural selection, in his polemical paper Zur Abwehr des ethischen,
des so'^ialen, des politischen Danvinismus, wherein the old student of evolu-
tion sharply criticizes the outgrowths that Darwinism had induced in the
sphere of social life. The fact is that a number of writers, partly biolo-
gists with a deficient social grounding, partly newspaper-men and political
authors of various kinds, had made use of the theory of the struggle for
existence and of selection to proclaim a new social theory, on the one hand
rejecting activities based on Christian charity and strivings after social
equality, and on the other hand extolling war, social want, and ruthless
competition as phenomena destined to thin out the weaker and less hardy
human beings, and thereby to further human progress. Against these asser-
tions Hertwig maintains that natural phenomena cannot be made the stand-
ards of human culture; justice and morality have their origin exclusively
in human community life; in nature no such principles exist; the beast of
prey that tears its victim to pieces acts neither justly nor unjustly, but ac-
cording to its nature. To make a merciless struggle for existence the basis
of social life would therefore be equivalent to destroying all that the cul-
tural efforts of the past have built up. War and economic misery are of no
constructive value; on the contrary they ruthlessly destroy both the capable
and the incapable. Thus in the very bitterest days of the Great War Hertwig
dares to hope for a peaceful settlement between the nations. That, however,
he did not live to see; when he died, in i9xx, the unhappy consequences of
the war — unhappy for the whole of humanity and most of all for his own
fatherland — had been brought out in all their frightful clearness, and his
570 THE HISTORY OF BIOLOGY
distress at these misfortunes is said to have broken the aged patriot and
philanthropist and to have appreciably shortened his life.
Eimer's "orthogenesis" theory
We shall mention in the following some further tests of the descent theory,
though more as proofs of the increasing difficulties with which the successors
of Darwin had to contend than for any actual value that the results possessed.
Theodor Eimer (1843-98), a fellow-countryman and disciple of Kolliker's
and eventually professor at Tubingen, endeavoured to solve the difficulty of
the descent theory chiefly along the lines laid down by Nageli. He rejects
Darwin's theory of variation in all possible directions as the basis of selec-
tion; the development of the organic forms must rather, he holds, depend
upon a force operating in a definite direction, a force induced and modified
by outward influence, such as light, air, heat, nourishment, and thus one
that provides selection with the material for changes which it influences.
This definitely directed evolution he terms "orthogenesis" and he seeks
to prove that it is indispensable for the building up of species; selection
alone cannot produce anything new, but rather it is this inner force, law-
bound, yet aff"ected by external influences, that is the true origin of life-
forms. In proof of this he cites a large number of observations dealing with
the colour changes in butterflies, as well as the development of the skeleton
of vertebrates, which attracted great attention at the time and brought their
originator a large following, but which are now no longer up to date.
Semon s " mneme" theory
Richard Semon (i 859-1919), a pupil of Haeckel and at one time a professor
at Jena, pursued another line of thought. His theory of evolution is also
based entirely on the belief in the transmission of acquired characters, but
he endeavours to give this theory a direction more suited to the time by
submitting it in a new form. To his mind, the weakness underlying previous
theories of this kind had been due to lack of clearness in the term "char-
acter"; instead of transmission of acquired characters it should be called
transmission of acquired reactions. For the hereditary transmission depends
upon the nature of the germinal plasm, and this reacts under natural laws
to the influence of the general condition of the body. The power of living
substance to react, its " Rei^barkeit," is the primary cause of evolution. Even
if the external influence upon it is transitory, there nevertheless remains an
impression, which becomes a decisive factor in its future development. This
impression is termed " Engramm." And the altered construction of the new
generations' form, upon which natural selection has an extenuating influence,
is due to the co-operation between the outwardly induced Engramm of the
body and that o the germinal plasm. This influence of the corporeal sub-
stance upon the germinal plasm is called "somatic induction," and the as-
sumed power of the living substance to preserve external impressions, upon
MODERN BIOLOGY 571
which the whole theory rests, is called " fnneme.'' Semon adduces a mass of
arguments to prove his theory of the transmission of external influence, but
all of them have since been rejected or given a fresh interpretation by the
representatives of experimental heredity-research. The purely evolutional
proofs are borrowed partly from palaeontology — for example, the process
of stunted growth in the toes of certain animal forms — and partly from
embryology — • the skin of the human sole is even in the embryonic stage
thicker than that on other parts of the body — but as these proofs cannot
possibly be tested as regards the evolution that has actually taken place,
they cannot be considered binding according to exact methods. The experi-
mental proofs, however, will be more closely dealt with later on; as a matter
of fact, they recur quite regularly in all Lamarckists and are adduced by them
as being positive, whereas other investigators have pointed out fallacies
either in the experiments themselves or in their interpretations.
Vauly s Lamar ckism
We must mention only in passing one more line of thought based on La-
marck, represented, inter alia, by August Pauly (1850-1914), professor at
Munich, who seeks the cause of evolution in a conscious psychic striving
towards a certain goal on the part of the organism and all its elements.
This theory can hardly come within the scope of natural science; it has
crossed the border of metaphysics, but is worth mentioning as a further ex-
ample of the desperate expedients that the speculation on the problem of
origin was finally compelled to adopt. In support of his views Pauly further
cites the traditional examples of the inheritance of acquired characters.
The struggle between the champions and the opponents of the theory
of the inheritance of acquired characters is to some extent reminiscent of
that between Pasteur and Pouchet regarding spontaneous generation —
people could not agree upon the interpretation of the experiments and a
theoretical standpoint based on them. And in this case, too, it would seem
as if the practical consequences must eventually determine the value of the
theory; neither animal-breeders nor horticulturalists have really succeeded
in applying the theory of the transmission of acquired characters, those of
them who believed in it having really remained at the same grossly empiri-
cal standpoint as their colleagues had adopted long before Darwin's time,
while modern racial research, working on the Mendelian method, has at-
tained results of an entirely different practical value and has, in fact, com-
pletely revolutionized the methods of racial breeding.
Plate s Darwinism
While both the theory of selection and Lamarckism thus had their sup-
porters, who carried the conclusions of their several lines of thought to ex-
treme limits, the middle course pursued by Darwin himself has also been
xoUowed by scientists of repute up to modern times. As one of these may be
572. THE HISTORY OF BIOLOGY
mentioned Ludwig Plate (born in i86i), Haeckel's disciple and successor
at Jena. After having applied his theoretical ideas to a number of special
investigations, both morphological and experimental, he summarized his
main arguments in an extensive work, Selektionsprinxif und Probleme der Art-
bildung, which may be said to contain all that can be adduced in modern
times in defence of the old Darwinism. And as its champion Plate has done
a great service, thanks to his wealth of knowledge, his strong convictions,
and his honesty. From the imperious disposition of his master, Haeckel,
he kept entirely free; he refused to employ the theory of selection to explain
the fundamental qualities of the living substance: assimilation, growth, res-
piration, etc., or indeed to explain variability or heredity; its sole function,
to his mind, is "to explain the origin of the teleological organizations, in
so far as they are not elementary qualities nor can be placed in the category
of Lamarckian factors." Darwin's greatest service, in his opinion, lies in
the fact that "he sought to explain organic finality out of natural forces,
to the exclusion of any metaphysical principle operating with conscious in-
telligence." Finality is thus the principal quality of organic life; adaptations
are expressly declared to represent a main difference between animate and
inanimate bodies. In his anxiety to defend the theory of selection, Plate
has, obviously without realizing it, hereby come perilously near Johannes
Miiller's old doctrine of finality and a far cry from Haeckel, who was at
one time prepared to characterize rudimentary organs as being the opposite
to profitable^ and the knowledge of them as "dysteleology." Plate, how-
ever, examines all the different kinds of finality — phenomena of correla-
tion, mechanical equiponderant apparatus, embryonic structures, instincts,
protective resemblance, and a good deal more — all this in order to find
proofs of the operation of natural selection. But if the theory of selection
is thus to stand or fall by the question of finality in nature, the result will,
of course, be that the function of selection automatically lapses if a different
view of natural phenomena is advanced. And, as has already been pointed
out, ever since the days of Democritus of old, research has constantly aimed
at seeking the existence of law-bound necessity in nature without any ex-
planations of purpose. But then, as Johannsen says, finality in an organism
becomes merely an expression for the fact that "organisms must be systems
in dynamic equipoise," that finality in general is self-evident in the very
fact of organization. From this standpoint one does not, of course, explain
by external causes the origin of functional adaptation in the organisms,
^ Haeckel has made a special point of the human appendix as a proof of nature's lack of
finality; it serves no purpose, but produces only dangerous inflammations — an extremely in-
genuous argument. A healthy appendix, of course, plays its part in the renewal of substance,
and the danger of sometimes becoming inflamed is one to which any section of the intestine
whatever is exposed.
MODERN BIOLOGY 573
which differentiates them from inanimate natural objects; on the contrary,
this becomes part of the problem of life itself, but instead we get rid of the
anthropomorphistically childish speculations upon teleological and non-
teleological forms of organizations in nature, which actually imply a con-
fession that we cannot grasp the idea of nature's being law-bound.
In such circumstances it is of no particular interest to follow Plate in
his attempts to meet all conceivable objections to the theory of selection.
Among other things he honestly admits that it has never been possible
actually to observe selection, but he consoles himself with the numerous
indirect arguments that could be quoted in its defence. Furthermore, he
emphatically maintains that, besides selection, environment co-operates in
renewing the organisms and their descendants; he thus rejects Weismann's
theory of the omnipotence of selection, as well as Elmer's and other neo-
Lamarckists' denial of its metabolistic power. He also tried to adopt an
attitude, consistent with his own point of view, towards the results of
experimental heredity-research, dealing with them in a monographic Verer-
bungslehre.
With this we can leave the doctrine of descent in the old Darwanistic
sense. Modern heredity-research has introduced quite a different and essen-
tially experimental treatment of the problems of evolution, and the old
morphological speculation upon the origin of species and genera has pro-
portionately lost ground — as it has always happened in the history of the
exact sciences that speculation must give way to facts. The old doctrine of
descent actually possessed the weakness of insisting upon external grounds
of explanation for the phenomena of life; selection as well as direct outward
influence were to explain phenomena that must really be an expression for
manifestations of life itself. Nevertheless there appears here also an increas-
ing self-deliberation based on a progressive knowledge of facts; there is a
vast difference between Haeckel's belief in the power of Darwinism to ex-
plain anything whatsoever and Plate's modest limitation of the function of
selection to a mere explanation of the teleological arrangements of living
creatures, "in so far as they are not elementary qualities." Indeed, why not
let them all be elementary qualities? One can trace here the human age-old
tendency to look for outward causes for everything; formerly one sought in
the phenomena of life manifestations of a divine creator; when this was no
longer perceivable, one had to look for a material creative power — it was
difficult to realize that evolution is a part of life itself. We shall revert below
to the problems of evolution in the form in which modern heredity-research
has presented them.
CHAPTER XVII
EXPERIMENTAL BIOLOGY
I.
Experimental Morphology
THE HISTORY OF BIOLOGY might really close with the establishing of
the dissolution of Darwinism. This theory undeniably constitutes
a construction of thought of its own, and the ideas and methods
of research that have succeeded it are still in the developing stage, and their
possibilities of development can at best be only guessed at. A summary
glance at these conquests of the newest biology is nevertheless defensible
as a further justification for the defeat of the old ideas and in view of the
intrinsic interest that the new discoveries possess. The author, who himself
laboured exclusively in the sphere of the old morphology and who accord-
ingly does not feel justified in competing with the many splendid presen-
tations of the development of experimental biology that already exist,
proposes to make only very brief reference to the results of modern research
and give a short account of the theoretical reflections to which they have
given and may give rise.
In a lecture given in 1900 in celebration of the birth of the twentieth
century, Oscar Hertwig gave a summary of the history of biology during
the past century. In it he sharply criticized physiology, which, in his view,
had created a brilliant experimental technique and had discovered a number
of facts concerning the chemical and physical processes of the organisms,
but, on the other hand, had neglected all other vital phenomena, with the
result that a whole category of the most important physiological problems
— namely, fertilization and embryonic development — had fallen entirely
into the hands of the morphologists and been dealt with by anatomists,
zoologists, and botanists. While the professional physiologists had thus re-
verted to "an empty mechanism" and imagined that the explanation of life
was only a chemico-physical problem, the morphologists discovered the
structural conditions in the fundamental elements of the body that are char-
acteristic of the phenomena of life and thereby extended the knowledge of
life in a sphere in which the methods of chemistry and physics have no part.
This accusation against the old classical physiology has certainly been to
some extent justified and is indeed confirmed by a statement made by one of
574
MODERN BIOLOGY 575
its representatives already quoted above; owing to this exclusiveness physi-
ology came to be an antithesis to morphology, which proved disastrous,
not least because the experimental method and the physiological point of
view no longer served the purposes of morphology, which in consequence
reverted to narrow phylogenetical speculations. Botany was the first to free
itself from this narrowness of view; indeed, there appeared at quite an early
stage students of this subject who were capable of not only realizing but
also solving physiological problems.
Julius Sachs was born in 1831 at Breslau, of poor parents. His studies
at school were embittered by privation and could be continued only thanks
to the kindness shown him by the aged Purkinje, whose sons were his school-
fellows. In their home he found help and encouragement, especially in his
interest in botany, which he displayed at an early age. When Purkinje moved
to Prague, young Sachs's prospects looked gloomy, especially as he was now
an orphan, but fortunately he was not forgotten by his old benefactor; he
was allowed to go to Prague after him and to work in his institute as an
assistant and draughtsman, while he completed his studies. Having received
his degree, he was given a post as teacher of botany at the Saxon academy
of forestry at Tharand, afterwards obtaining a similar situation at Bonn and
finally being called in 1868 to Wiirzburg, where he laboured for nearly thirty
years, gaining a brilliant reputation both as an investigator and as a teacher.
In his best days Wiirzburg was an international centre for botanical re-
search. Towards the close of his life his powers waned, and at the same time
he lost his ability to follow the development of evolution; his self-conceit
had always found it difficult to keep within reasonable bounds, and in his
old age he simply could not endure any other opinion than his own. This
eventually resulted in isolation, which embittered his existence. He died in
1897.
Sachs was one of those who was early won over to Darwinism, and
throughout his life he viewed biology from the angle of the doctrine of
descent. This was in fact the reason why in his otherwise praiseworthy
Geschichte der Botanik he speaks so contemptuously of Linnasus, who to him
was conspicuous only as a narrow-minded apostle of the constancy of species.
But Nageli has also exercised a great influence upon Sachs, who, for instance,
associates himself with a view that there are internal causes of form-develop-
ment in living creatures, and he also embraced the theory of protoplasm's
being composed of solid particles capable of absorbing water in their inter-
vening spaces. In regard to heredity, Sachs holds views most closely reminis-
cent of Weismann's germinal-plasm theory.
Sachs creates experimental plant-biology
Sachs, however, is best known as the creator of experimental plant-biology.
This science had really made but little progress since the days of Saussure,
576 THE HISTORY OF BIOLOGY
for both the botanists of the romantic school and their immediate successors
had mostly applied themselves to morphology. Sachs, on the other hand,
conscientiously devoted himself from the beginning to the problem of plant
physiology and the means of exploring it. Among his earliest works may
be mentioned his investigations into the physiological part played by the
chlorophyll granules, which he elucidated in its most essential features: he
established the fact that starch formation is the first product of carbonic-acid
assimilation, and, further, that sunlight plays the decisive part in this proc-
ess. He investigated the different parts of the solar spectrum with the view
to discovering their influence upon the alimentation of plants. Again, the
continued metabolism and conveyance of nutrient substances within the
plant has been worked out by him in all essential features. He studied all
the vital processes in the vegetable kingdom with a view to ascertaining
their intensity, which he expressed in graphic form. It was mainly, however,
the movements of flowers that he systematically investigated; he established
the dependence of the direction of growth upon the law of gravity and in-
vented a rotating apparatus by means of which this growth can be studied
under abnormal conditions. He is responsible for the method of investiga-
tion and thorough exploration of all that goes by the name of tropisms —
at least as far as the vegetable kingdom is concerned. As a basis for all these
phenomena he mentions irritability, a quality which he believes originates
only in the living plasm and which he carefully analyses in respect of its
various manifestations — an investigation that has been of immense theo-
retical importance for subsequent research. Finally, it may be mentioned
that Sachs was the finest text-book writer of his time. Through his manuals
the facts that he discovered and the ideas that he defended have spread
far beyond the circle of his personal pupils.
Among these pupils Wilhelm Pfeffer (1845-1910) is worthy of spe-
cial mention. When still quite young he became a professor, first at Tubingen
and afterwards at Leipzig, where he subsequently laboured as a brilliant
teacher and investigator. He made a special study of the phenomena of
growth in plants and the influence exerted upon them by both external and
internal factors. He investigated a number of external influences and made
valuable contributions to our knowledge of them, at the same time consid-
erably improving the technique employed for their study. However, he
emphatically declares that external influences do not act directly in a de-
velopmental way, but only cause a change of activity in the plant itself.
He sees in the study of this reciprocal action between external influences
and internal manifestations of life the very aim of biology, and '' dkjormative
Determinierung der Zellen und der Organe ' ' becomes the object of his close analy-
sis. In this study very careful attention is paid to all co-operating factors,
so far as they can be calculated, and unwarranted attempts at simplification
MODERN BIOLOGY 577
are rejected; it is thus definitely declared that cell-division is a physiological
process and not merely a question of increased superficial distention within
the cell, as mechanistically inclined investigators have tried to explain the
phenomenon. Generally speaking, clear traces of PfefTer's influence as re-
gards the presentation of problems and general points of view are also to
be found in experimental zoology, a fact, indeed, that some of its exponents
have openly acknowledged. His labours have proved of still greater impor-
tance to botanical specialized research-work; he is generally acknowledged
as one of the leading personalities in botany in modern times.
Another eminent pupil of Sachs was Karl Eberhard Goebel (born in
1855), "^^° ^^^^ ^^ ^^^ '-^'^^ ^^ assistant at Wiirzburg and has since worked
as a professor at several universities, latterly at Munich. He is an investi-
gator of many parts, being a plant-geographist with a wide experience of
the tropics, a morphologist, and a physiologist. As a morphologist he has
always maintained the dependence of form upon function; morphology
should no longer be kept separate from physiology, as formerly. He employed
the old word "metamorphosis" to denote organic development, but not
in the old idealistic sense; by it he would express the idea that a change in
function produces a change in form. "Our idea of metamorphosis is thus in
the first instance ontogenetical, and thereby experimentally conceivable and
demonstrable," he says. According to him, no "indifferent rudiments"
exist, for every rudiment has its peculiar qualities, which determine its de-
velopment, and this can suffer change only as the result of definite changes
in the vital manifestations. "If we call a leaf-rudiment at any stage 'in-
different,' it really means nothing but a denial of the causal connexion of
evolutional phenomena." These changes in the conditions of life can be pro-
duced experimentally, and extremely important experiences in regard to the
organic construction of plants can be gained thereby. Space forbids our going
closely into Goebel's experimental studies in " Or gano graphic,'' as he calls
the study of the relation between an organic form and function. He has
hereby performed a great service in furthering the investigation of plant
evolution and the mechanical process in evolution in general.
Even apart from Sachs's school there have been many important stu-
dents of plant biology who have produced valuable results; a couple of them
may be mentioned here. Julius Wiesner (i 838-191 6), professor at Vienna,
has carried out useful experiments in the sphere of technical botany, espe-
cially in regard to the effect of light upon the vegetable world, which he
studied in different localities and under different experimental conditions.
Hans Karl Albert Winkler (born in 1877), professor at Hamburg, brought
to light the curious graft-hybrid phenomena: as a result of grafting related
plant-species their tissues can be made to grow through one another in dif-
ferent ways.
578 THE HISTORY OF BIOLOGY
In zoology the experimental method has made slow but sure progress.
Strictly speaking, its development has gone hand in hand with the new con-
ception of the fundamental problems of biology that has gradually usurped
the place of the old phylogenetical idea, dating from the zenith of Darwin-
ism. It may be said that the new principle had already been enunciated by
Kleinenberg in his above-quoted saying that an organ's form depends upon
its function and not upon its origin. Almost at the same time as this utter-
ance was made, August Rauber (i84z-i9I7), professor of anatomy at Dorpat,
sought to discover the conditions and laws governing the first construc-
tion of form in the vertebrate embryo. In opposition to the contemporary
belief in the independence of the individual cells he maintains that the whole
governs the parts and not vice versa; the egg-cell determines the directions
of cleavage by its division of matter, growth is the primary and cleavage
the secondary. Division into many cells facilitates the metabolism of sub-
stance and renders possible greater strength in the organism and, by special-
izing the elements, a far more extensive division of labour, but even when
this division of labour is at its maximum, the organism remains a whole,
the parts of which are developed under the influence of the whole. Through
his efforts to find out in detail the conditions governing the various phases
of development, Rauber became, along with His, Goette, and Kleinenberg,
a precursor of the later school of evolutional physiology.
As its founder is named by universal accord Wilhelm Roux (1850-1914).
He was born at Jena, where his father was a fencing-master. He studied
first under Haeckel and then, at Strassburg, under Goette, and also in Berlin.
He became professor, first at Innsbruck, then at Halle, where he worked
during the period 1895-19x1. He laboured with never-failing energy and
powers of endurance, in speeches and in writing, as a teacher and an agita-
tor, on behalf of the method of research and the line of investigation that
he originated. As a research -worker he has already been outdistanced by
younger minds, but he will always be regarded as a pioneer. However, the
same line of thought has been followed from other quarters as well — as,
for instance, by the disciples of the above-mentioned plant-physiologists,
Sachs and Pfeffer, whose ideas were really in many respects in accord with
those of zoological evolutional mechanics.
Roux was a pupil of both Haeckel and Goette; his works, in fact, bear
traces of the influence of both, not only in his early days, but even at a far
later age: even into the nineties phylogeny still represented the aim of his
research work, and the struggle for existence and selection appear to him
the most vital forces of life. But he would achieve this aim, not like Haeckel
through a mere comparison between more primitive and more developed
forms, but through investigating the mechanical process in ontogenetical ev-
olution, such as through the program that Goette had in mind. As a matter
MODERNBIOLOGY 5 79
of fact, Roux also has points of contact with Weismann, whose theory
of the continuity of the germinal plasm he embraces; he consequently re-
jects the theory of the heredity of acquired characters. He resembles Weis-
mann, too, in the fact that he likes discussing hypothetical entities of life,
whereof he enumerates a whole series, which, however, it is not worth
while dilating upon here. For these very reasons he was on bad terms with
O. Hertwig, who, as we have seen, entertained quite different views, just
as, on the other hand, he was at variance with Haeckel and other original
Darwinists on account of his mechanical theory of evolution.
Creation of evolutional mechanics
The theoretical speculations upon descent are, however, a less essential side
of Roux's research work. He will mostly be remembered as the creator of
experimental embryology; whether the theories that he based upon his ex-
periments are eventually accepted or rejected, there is no doubt at any rate
about the fact that he created a special method of research, which has proved
productive and was largely employed by his contemporaries. But he also
exercised considerable influence upon the theoretical conception of biology
itself; he has directed research to a series of problems which many, following
his precedent, have taken up for treatment and which have largely guided
modern biological research. He himself defined as the aim of the new science
that he desired to found the elucidation of the "true causes of formation"
to which all living creatures and every single individual must attribute
their origin — a subject in which the earlier "descriptive natural science,"
to his mind, failed to show any interest. These causes of "form-building
forces," whereby the individual organism receives step by step the form that
characterizes it, must, in his opinion, be studied primarily by way of ex-
periment; if a process of development is altered by different kinds of inter-
ference, it is possible by combining the results to discover the cause of the
process; thus is created a "causal-analytical form of research," such as chem-
istry and physics had already realized in many instances.
Natural selection ivithin the organism
As previously mentioned, Roux began his activities with a theory of func-
tional adaptation produced by means of natural selection within the organ-
ism. This theory he sought to apply to the organs of various vertebrate
animals; he measured a large number of muscles in man and tried to deter-
mine to what extent their dimensions are dependent upon one another; he
studied the caudal fin of the dolphin from a mechanical point of view and
sought to determine the lines and curves in which these tissues are arranged
in order mechanically to sustain the function of the whole. After a short
time, however, he went over entirely to embryology, and in this field pro-
pounded the question to what extent and how far back in evolution certain or-
gans and tissues are predestined to assume their prospective form and function.
580 THE HISTORY OF BIOLOGY
and whether this predestination can be affected by different influences.
This problem, in fact, is the old antithesis of preformation and epigenesis
formulated in a different way, and the discussion of the problem has to a
great extent been carried on in the new form brought about by the employ-
ment of the methods of experimental embryology. Roux himself, starting
from the theory of the continuity of the germinal plasm, which was in-
fluenced by Weismann and Sachs, maintained the idea of a very early deter-
mination of the various parts of the embryo; in his view, the first cleavage
plane of the germinal egg establishes the median plane in the individual,
the cleavage furrow is determined by the line of penetration of the sperm,
and the front part of the future animal is already determined before the fer-
tilization of the egg through the amassing of cytoplasm in that quarter.
Each successive cleavage delimits a prospective part of the embryo; be-
cause in karyokinesis the substance becomes not equally apportioned to the
daughter-nuclei, these latter acquire different values, so that, as he says, the
whole process of development becomes a piece of mosaic work. In proof of
his theory Roux carried out a series of experiments, which excited universal
admiration at the time; he treated a newly-fertilized frog's egg in such a
way that with a heated needle he burnt away one of the two first-formed
blastomeres; then there developed at first a half-embryo, which afterwards re-
generated in the usual manner of the Amphibia. Roux performed many other
experiments with the same purpose in view, and though his technique was
simple as compared with that which his successors afterwards elaborated,
it was nevertheless he who first systematized this kind of research work.
Roux, however, was at once subjected to sharp criticism; O. Hertwig
in particular, who from the very beginning had been an opponent of
the Weismann heredity-theory and its founders, at once attacked the
"mosaic" theory, maintaining that the different parts of the egg are
by no means predetermined, but that, on the contrary, the egg's mass
is equipotential-isotopic, as he calls it. Moreover, Hertwig, who in-
variably opposed narrow mechanistic tendencies in biology, strongly ob-
jected to the actual term "developmental mechanics," which appeared to
him to imply an inadmissible schematizing of the phenomena of life. And,
finally, he considered Roux's experiments to be inaccurate; he imitated them
and found that the halved egg gave rise, not to a half-embryo, but to a
whole embryo of small size. Hertwig found immediate support in Driesch,
an observer who will be mentioned further in another connexion. He halved
newly-segmented eggs of the sea-urchin and obtained from each half one
larva half the size of the normal. On the basis of these and other experiments
he propounded a special theory of evolution, according to which each part
of the egg has, on the one hand, a "prospective value," and, on the other
hand, a "prospective potentiality"; the meaning of these terms will best
MODERN BIOLOGY 581
be explained by an example: the half-egg of the sea-urchin has the pro-
spective value of forming half a larva, but the potentiality to form a whole
larva; the former is thus the determination of the part that holds good in
normal cases, the latter the power of every part to compensate for another
if necessary. On the other hand, Roux found support in the American Wilson.
He isolated cleavage cells from the egg of a mollusc and found that they
became what they would have become in their normal connexion, and noth-
ing more. And later Boveri discovered that the sea-urchin's eggs investigated
by Driesch actually possess a differentiation that from the beginning deter-
mines the succeeding developmental orientation. The fact is, the disputants
have gradually had to reconcile their views; Roux had to abandon his theory
of the different-shaped cleavage of the nuclei, while Driesch had to give up
his theory of the absolutely uniform character of the sea-urchin's egg. On
the whole, these experimental discoveries were too generalized; what was
discovered in the case of one animal's egg was applied without question to
the entire animal kingdom, whereas in actual fact a vast number of different
conditions prevail. In the main, however, it seems as if subsequent research
has obtained results that would indicate generally that a very early special-
ization and localization of the rudiments take place in the embryo. The
investigations that have been carried out especially by Hans Spemann, one
of the foremost champions of experimental morphology at the present time,
give some evidence of this. Born in 1869 at Stuttgart, he studied at Heidel-
berg, Munich, and Wiirzburg, and has been a professor at Rostock and
at the Kaiser Wilhelm Institute in Berlin; he is now working at Freiburg.
An extremely clever experimenter, he has specially taken upon himself to
prove the determination of different sections of the embryo by transferring
parts of the body of a batrachian embryo from one place to another and
then studying the successive development. As a general result he records that
the main organs of the amphibian embryo are definitely determined during
the process of gastrula formation; at an earlier period, in the blastula stage,
normally formed twins can be produced by means of dividing off with thread;
at the beginning of the gastrula stage the forward end can still be doubled
by means of binding, but after that the possibility of regulation decreases,
and disappears altogether when the gastrula is completely formed. We must
here pass over the details of the minutely planned and carefully carried-out
experiments by which these statements are proved, as also the particulars
of the various attempts he made to get mechanical influences to operate —
pressure, binding up, and such means, whereby different observers have
sought to trace out the forces operating inside the egg and to discover the
details of which they are composed.
Besides these purely mechanical experiments others have, of course, been
carried out, such as those in which eggs and embryos have been subjected
58x THE HISTORY OF BIOLOGY
to the influence of electricity, light, heat, chemical compounds; of these
the last in particular have produced results of great interest. Among such
may be quoted the experiments of Curt Herbst (born in 1866), Biitschli's
successor at Heidelberg. He placed sea-urchins' eggs in various saline solu-
tions; in lime-free sea-water the eggs disintegrate into their first cleavage
cells, w^hich give rise to dwarf larvas; the addition of lithium salt produces
quite abnormally developed larvas; treatment with sulphate-free sea-water
also causes peculiar malformations in various parts of the egg. Undoubtedly
the most remarkable of these chemical experiments in evolution are, how-
ever, those concerned with the initial development of the egg. In this field
Jacques Loeb not only has taken the initiative, but has also proved a leader.
Born in Germany of Jewish parents in 1859, he studied medicine in his na-
tive country, becoming a doctor and assistant lecturer in physiology at
Wiirzburg, but he soon migrated to America, where he held a number of
professorships, the last one being at the Rockefeller Institute in New York.^
Already, in the eighties, R. Hertwig had observed that unfertilized sea-
urchins' eggs, if treated with a weak solution of strychnine, surround them-
selves with a membrane similar to that which appears after fertilization,
before the cleavage begins. Similar observations were made later by Morgan,
the well-known student of heredity. Loeb took up this problem for system-
atic revision and achieved results that at once attracted great attention and
in some directions produced very far-reaching conclusions and awakened
high expectations. After performing a series of experiments he worked out
a method of bringing the sea-urchin's eggs to the larval stage without ferti-
lization. This method is somewhat complicated; first of all, the eggs are
subjected to the influence of a weak organic acid, which induces the forma- ,
tion of membrane, after which they are placed for a carefully fixed period
in sea-water, the salinity of which has been increased to one and one-half
times the normal and which besides has been mixed with soda, and finally
in normal sea-water, in which the development takes place. Several other
simpler methods have also produced results, but they have been few and
indefinite. The above method, however, subject to careful regulation, works
safely, although, contrary to Loeb's statements, the larv^ thus formed are
by no means invariably quite typical. On account of this, Loeb has tried to
ascertain the chemical compounds that induce development and has come
to the conclusion that the membrane formation is caused by the fat-dis-
solving capacity of the influencing acid, and that the spermatozoon, which
upon penetrating the egg has the same effect, must produce a similar sub-
stance. The "hypertonic" sea-water then used "corrects" this effect and
thereby contributes to the actual segmentation. Thus the multiplication of
^ Loeb died in 1924.
MODERN BIOLOGY 583
cells would be reduced to a relatively simple sequence of chemical processes.
There can be no doubt, however, that Loeb has simplified overmuch. Other
investigators have, in fact, produced the same developmental phenomena
by entirely different methods. Among those who have worked in this field
may be mentioned, apart from Loeb's own pupils, Yves Delage (1854-1910),
professor at Paris, who worked out his own method of developing the sea-
urchin's egg, and A. Bataillon, professor at Dijon, who acquired a name
especially as a result of his experiments with the frog's egg; by simply
pricking the egg with a needle dipped in serum, he caused the frog's eggs
to develop into larvas — a method that "activates" the egg in an entirely
different way from Loeb's. Here we obviously have a metabolistic process
within the very mass of the egg, set free by mechanical irritation — as a
matter of fact, the dose of serum has latterly proved to be superfluous —
although the adipose splitting can in this case also be established; it is here
induced by the egg's own vital manifestations and not by any solvent in-
troduced from outside. Loeb's theories, to which we shall revert, are gov-
erned entirely by his lack of interest in morphological phenomena and by
his consequent passion for schematizing the complex vital process. Strictly
speaking, most other evolutional physiologists of the new direction, from
however mechanical a point of view they may otherwise have regarded
evolution, have nevertheless divided its mechanical phenomena into those
having a purely external origin and those that result from specific internal
causes. Among the former are counted the universally observable influences
of heat, electricity, chemical reagents; among the latter, the various organ-
isms' peculiar ways of reacting to them, as well as to purely mechanical inter-
ferences, with their normal existence. In the course of studying these hetero-
geneous phenomena of reaction many of the experimental observers of our
own time have produced and are still producing a great number of detailed re-
sults of immense interest, achieved by the employment of exquisitely delicate
methods. For the details of this field of inquiry — still by no means ex-
hausted — we must refer the reader to technical literature on the subject; a
number of general points of view that have arisen in the course of the work
carried out in connexion with these subjects will be discussed later on. We
shall instead pass on to another form of experimental research — a field that
is without doubt of the greatest interest to humanity at the present time.
■L. Experimental Heredity-research
Earlier ideas on heredity
"Inheritance" and "heredity" are terms that originally belonged to the
judiciary and have been borrowed from it to acquire a natural meaning.
584 THE HISTORY OF BIOLOGY
Just as the right of a child to take over his parents' property is called in-
heritance, so the word is used to denote the fact, which has been known of
old, that children resemble their parents in body and soul; facial features
and figure are said to "be hereditary ' ' from father and mother to son and
daughter. It can be no matter for surprise that from the beginning the mean-
ing of heredity in the biological sense has been considered to be this: the
direct transmission of qualities from parents to children — that, in fact,
has been the idea up to recent times and it has not been until the details
of this transmission came to be studied that a deeper insight has been gained
into the true facts, and an entirely new conception has taken the place of
the old theory of transmission. It is through this research work that the
problems of evolution have for the first time been dealt with on an entirely
exact basis; the same mathematical exactness that formerly only experi-
mental physiology was capable of achieving now characterizes the methods
and results of heredity research.
Exact heredity-research has received contributions from various quar-
ters. Investigators with a Darwinistic training have made weighty contri-
butions to it, but besides these, others — and, in fact, the most valuable
of all — have come from circles that have had nothing whatever to do
with Darwinism. In a previous section have been mentioned the investiga-
tions into the hybridization of plants that were carried out in the eighteenth
century by Koelreuter. His experiments were taken up by many other stu-
dents, among the most highly reputed of whom may be named Karl Fried-
rich Gartner (1772.-1850), a medical practitioner by profession, whose
elaborate experiments with plant hybrids brought him a great reputation
and were especially taken advantage of by Darwin. The experiments carried
out by the Frenchman Louis Leveque de Vilmorin (18x6-60) were of a
different type. De Vilmorin belonged to a family that for generations had
carried on trade in grain and seed-cultivation; he himself was particularly
interested in sugar-beet, which he cultivated with a view to increasing its
percentage of sugar. In this he started from the principle that the offspring
of each individual should always be kept separate; he collected the seeds
of beets with a high sugar-content and sowed them separately, with the
result that he obtained cultures of a very valuable quality; in doing so he
discovered that individuals that look alike might have entirely different
characters, and he thus came to hold the view that the power of inheriting
characters might in itself vary and give rise to heterogeneous offspring.
Through these results de Vilmorin became a pioneer of modern heredity-
research, and his theses, based, as they are, upon exact observations, possess
quite a different value from his contemporaries' "philosophical" specula-
tions on heredity, numerous traces of which are to be found, inter alia, in
belles-lettres of a naturalistic type.
MODERNS lOLOGY 585
The problem of heredity was dealt with from an entirely dilTerent
point of view by Francis Galton (i82.i-i9ii). He was a cousin of Darwin's
and his life reminds one of the latter's. The son of wealthy parents, he
studied medicine, but never graduated, for as soon as he had inherited his
father's fortune, he went on voyages, first to Eygpt and then to south-west
Africa, where he made valuable geographical discoveries. After his return
to England he applied himself for some years to meteorology, but eventually
devoted his life entirely to heredity research. He began by seeking to prove
experimentally Darwin's pangenesis theory, which, it will be remembered,
assumes that particles from all the organs of the body are transmitted
through the sexual products from generation to generation and bring about
the offspring's resemblance to the parents. Galton injected blood from a
foreign rabbit into a pair of grey ones, in the hope that their progeny would
thereby become dappled, which would have been a proof of the pangenesis
theory; but his expectations were not fulfilled, the young of the grey rabbits
turning out grey. In virtue of this experience Galton produced a heredity
theory of his own: all the organic units existing in a fertilized egg he terms
"stirp," and he believes that the majority of these are used for purposes of
organic structure, but that a number of them are left over, and these give
rise to the sexual cells, the qualities of which are thus not influenced by
the conditions of life of the individual. He gave expression to this opposi-
tion to the doctrine of the heredity of acquired characters as early as in
1875 — ^^^^ ^^' before Weismann — but after that he passed on to quite
different developmental problems. From the very beginning the evolution
of man had been a subject of special interest to him; he is extremely fond
of demonstrating physiological phenomena with the aid of social and politi-
cal comparisons. The fact that brothers and sisters are often so unlike one
another is explained by a reference to an electoral body, where a very slight
difference of opinion can often give rise to an entirely different result when
it comes to the vote; in the same way, a slight change in the hereditary sub-
stance can produce a complete change in the appearance of the individual.
Galton s statistical method
In a later work. Natural Inheritance, Galton has presented the statistical
heredity-theory that has made his name famous; his stirp theory is here
abandoned and the transmission of acquired characters is no longer denied
so absolutely as it had been before. Instead, basing his arguments on an
exhaustive research-material, he tries to find out the variational direction
in a large number of human qualities. The most highly valued of his investi-
gations have been those into the question of height; the height of a number
of parents was compared with that of their grown-up children arranged in
series from the shortest to the tallest; there is thus obtained an average
height, which is possessed by the majority, while the less numerous
586 THE HISTORY OF BIOLOGY
extremes of abnormally short and tall persons group themselves in either di-
rection from the middle. Now, Galton finds as a result of his heredity research
a variation towards the mean value, the children of the tallest groups of
parents having become tall, but not so tall as the parents; those of the short-
est parents, on the other hand, having become short, but not quite so short
as the parents themselves. From this Galton infers that there is a heredity
variability in a certain direction, which natural selection, of course, influ-
ences. This result, seeing that it first came to light when Darwinism was at
the height of its popularity, was bound to win many supporters, as also did
Galton's principle that it is necessary to deal statistically with as large a
number of cases as possible, seeing that law-bound necessity in isolated cases
is effaced by incidental circumstances. This principle was bound to attract a
generation that preferred to regard humanity collectively and placed but little
value on what was purely individual. Later, however, it has been found that
this collectivism actually constituted Galton's most serious weakness; it is
practically an impossibility to draw conclusions regarding the individual case
from statistical mass-calculations, and if it is attempted, it leads to absurd re-
sults. On the other hand, Galton's service lies in the fact that he introduced
exact measurements and mathematical calculations into the theory of evolu-
tion; his method of expressing the details of development graphically by
means of curves has since been applied with great success by students who
were able to isolate well-defined phenomena and to follow them up through
different generations. For Galton's chief weakness was really this, that he
believed that he could deal with practically anything statistically; he never
realized that an object which is to be examined must first have its true essence
determined. Thus, in his above-mentioned work he tried to determine statisti-
cally the laws governing "marriage selection" — inter alia, whether persons
of different dispositions feel attracted to their likes or vice versa. From one of
his tables it appears that 46% of married men are ill-tempered; of these, again,
2.7.% have had good and X4% bad-tempered wives. ^ Statistics of this sort
seem far more suited to a comic paper. The fact that Galton seriously tried
to solve such a problem testifies to his extraordinarily dilettante mind, which
cannot be excused by the fact that even at a later period an occasional stu-
dent of heredity has sought to ascertain the existence and transmission of
equally vague and indefinite human qualities. As a matter of fact, Galton
applied his method not only to human beings; he also experimented in hor-
ticulture, dealing with the results thus obtained by the same statistical
method. He bequeathed his fortune to an institute for heredity research in
London, which afterwards worked in accordance with the principles that
he had laid down. Galton had human welfare very much at heart; he wanted
^ See "Appendix D" in chat work.
MODERN BIOLOGY 587
to create a better human race and desired that all research work should serve
that object; he gave to the science that he placed highest of all the name
of "eugenics," a name that has become universally accepted.
A new phase in the history of heredity research was introduced by Hugo
DE Vries. Born at Haarlem in 1848, he studied at Wiirzburg under Sachs,
held various posts in Germany, and finally became professor at Amsterdam.
He applied himself first to the study of plant physiology and published val-
uable results of his investigations into the pressure conditions in plant-cells.
At the same time he speculated over Darwinistic problems. He, too, pro-
duced a theory of life-entities, which he called "pangens," by which he
meant those qualities in the organism which are capable of independent
variation, and each of which must, in his view, be represented by one ma-
terial entity. Like so many other biologists of the younger generation, how-
ever, he entertained doubts as to the ability of the traditional Darwinism
to solve the problem of evolution. He was especially preoccupied with the
undeniable fact that the species in nature remain constant and that the slight
transitions whereby, according to the old theory, one species is converted
into another can never be observed; the species is a self-contained entity,
and yet the conversion of species must have taken place in the course of
the ages. Kolliker's previously mentioned theory of sudden changes of species
seemed to him to offer the possibility of adjusting this inconsistency, nor,
indeed, had Darwin himself denied the existence of sudden "single varia-
tions." It was only a question of obtaining actual proof of the existence
of such changes. Eventually de Vries believed that he had found it in CEno-
thera lamarckiana (evening primrose), a plant introduced from America, which
has spread over various European countries. This plant grew in masses in
a meadow in the neighbourhood of Amsterdam and exhibited, besides a
number of typical forms, some few with an entirely divergent appearance.
A number of specimens having been transferred to a garden and there al-
low^ed to multiply by self-fertilization, it was discovered that in the course
of a few years there developed out of seed of the old species not only forms
similar to it, but also isolated specimens With, well-marked new species-
characters, which were retained for the purpose of further cultivation:
among them a dwarf form, a markedly latifoliate form, and some others.
Here, then, we get, according to de Vries, a species that suddenly "ex-
ploded," as he expresses it, and gave rise to a number of new species, each
with definite characteristic features. This case shows, according to him, how
the species in general have arisen; the species, he says, are no arbitrary
groups, but completely independent entities, delimited in time and space,
which originate through old species' suddenly disintegrating into a number
of new forms; of these some are capable of life and survive unchanged until
the next mutation, while others cannot sustain the struggle for existence
588 THE HISTORY OF BIOLOGY
and succumb to natural selection. All fresh characters have thus been formed
as a result of mutations; between the mutations a species survives with its
characters unchanged; the slight variations that occur daily in the life of
the species have no effect on evolution, because they are not hereditary, and
the recombinations of characters that arise through the crossing of different
forms have no new significance. "As many steps as an organism has made
from the beginning, so many mutation periods must have occurred." These
mutation periods, de Vries believes, must have arisen at a far more rapid
pace during previous geological periods than they do nowadays, and he is
thus able to explain on the basis of this theory the origin of the forms of
life without the assumption of those infinite spaces in time that the old
Darwinism required at its disposal.
When it first appeared, de Vries's theory naturally met with violent
opposition on the part of loyal Darwinists. It was certainly admitted that
in all essentials he really accepted the point of view of the old Darwinism,
in that he maintained the theory of natural selection as the principle govern-
ing life, but the fact that he denied the heredity of the slight variations
and their importance for selection and maintained the immutability of species
in the normal existence between mutations was far too much at variance
with old traditional ideas to be acceptable. Every possible effort was made
to get away from the facts that he had adduced and the conclusions that
he drew therefrom. As a matter of fact, these certainly have been open to
objection. His cultural experiments with CEnothera have failed to withstand
the criticism of later years. Johannsen has objected that the material with
which the experiment was carried out was casually selected and was not
kept as pure as it should have been, and finally a Swedish naturalist, Heri-
BERT-NiLssoN, Carried out the entire experiment over again and came to the
conclusion that the new generations of Oenothera lamarckiana only show fresh
combinations of characters that already existed in the main species. De Vries,
who was one of those who rediscovered Mendel's law of cleavage, has, in
fact, denied the validity of that law as regards mutations, such as those of
CEnothera, but this has been found to be a mistake. Consequently his theory
of the formation of new species of that plant collapsed. His service to sci-
ence, on the other hand, lies in the fact that he revealed the phenomenon
of mutation, for that this phenomenon exists has since been proved over
and over again. Moreover, on the basis of his "pangen" theory he insisted
upon the necessity of analysing with regard to their elementary units the
hereditary qualities that characterize the species. "It is not a question," he
says, "of the origin of species, but of the development of the species-char-
acters." Through these assertions he became a pioneer of modern heredity-
research. "His mutation theory," Johannsen declares, "has represented the
principal milestone in the transition from the old ideas to the modern
MODERN BIOLOGY 589
conception of heredity and will always, therefore, retain its historical
significance."
WiLHELM LuDviG JoHANNSEN was bom in 1857 at Copenhagen, studied
there and in Germany, and eventually became a professor, first at the In-
stitute of Agriculture and afterwards at the University in his native country.
Being a pupil of PfefFer, he first of all went in for experiments in plant physi-
ology, making a number of interesting observations in this sphere in connex-
ion with the effect of ether on the metabolism and growth of plants. Soon,
however, he devoted himself exclusively to experimental heredity-research
and has gradually become one of the leading authorities in that field. Origi-
nally a supporter of Galton's statistical method, he quickly realized its
deficiencies: that by working with mixed m.aterial it reached conclusions
utterly at variance with the true facts. Starting from de Vries's insistence
upon the necessity of investigating the units of hereditary characters, he
began to follow with minute care the phenomena of heredity in generations
of plants. He purchased a quantity of beans, weighed them, and then culti-
vated the seeds of every bean separately; he thereupon found that within
a succession of individuals thus produced — a "pure line," as he called it —
there exists a certain type of hereditary units, which remains unchanged
throughout; if the plants are starved, both they and their seeds become small;
if they are manured, they grow strong, but this has no effect upon the heredi-
tary character; whether one sows small or large seeds of the same pure line,
one obtains under the same external conditions the same plant-type. There
is thus within one and the same pure line a certain hereditary type — a
"genotype," as it is called — that is unalterably the same, whether or not
the vital conditions alter the external form of the actual individual — that
is, the phenomenon-type, or "phenotype," as Johannsen called it. Those
characters which form the genotype are thus the only ones that are really
hereditary, whereas the phenomenon-type produced by environment has
nothing to do with heredity; stunted growth in generations of plants grow-
ing on poor soil is an external character, a "false heredity." There is there-
fore no possibility of acquired characters' being inherited, nor is there within
the pure lines any chance of variation of the kind assumed by Darwinism.
Those characters that go to make up the hereditary disposition Johannsen
terms hereditary factors, hereditary units, or genetic elements; in a pure
line, then, the genetic elements are the same: its individuals are homozygotes
("zygote" denotes the fusion-product of the male and female sexual cells),
while the offspring in life-forms that multiply by the pairing of different
individuals is a heterozygote, as it represents a fusion of the parents' vari-
ous inheritable factors. A heterozygous individual therefore always has a
hybrid nature, and special methods are necessary for the elucidation of its
qualities. But at this point we come to the subject of hybrid research, which
possesses a history of its own.
59° THE HISTORY OF BIOLOGY
JoHANN Mendel was born in i8iz of peasant parents at Heinzendorf,
a German colony in the midst of the Slav population of Austrian Silesia.
Having shown remarkable intelligence at an early age, he was sent to a
grammar-school, and, probably with a view to obtaining better opportun-
ities for devoting himself to study, he entered an Augustine monastery at
Briinn in the district of Moravia. As a monk he adopted, after the Catholic
custom, a new Christian name, Gregor, by which he became known to
posterity. He was sent at the expense of the monastery to Vienna, where
he studied for three years, devoting himself especially to mathematics and
natural science; upon returning home he became a schoolmaster and in his
leisure hours cultivated plants in the cloister garden for scientific purposes.
He published the account of his results in the little-known "treatises" that
were brought out by the natural-science society at Briinn. In 1868 he was
appointed head of the monastery, or prelate, as it was called. This appoint-
ment, however, actually proved his undoing. Four years later the then liberal
parliament in Austria sought to reduce the country's financial distress by,
inter alia, taxing the monasteries. The monks, like all the reactionary parties
in general in the country, considered that the tax menaced the monasteries'
ancient privileges and set themselves up in opposition to it. Eventually,
however, the measure was carried through in several instances, but the one
who refused to give in was Mendel; for twelve years he held out, defying
penalties and warrants of distraint, but finally he broke down completely
under the struggle, contracting a sickness that resulted in his death in 1884.
Thus fell one of the pioneers of modern biology as a champion of Catholic
clericalism — in its way an irony of fate.
Alendel's experiments with peas
Mendel's fame, which was late in coming, rests simply and solely upon
two short essays in the above-mentioned journal. They are, however, the
fruit of many years' work and testify to a keen observation of nature and
a thorough grounding in mathematical thought, which do not often go
together; Darwin, for instance, had a genius for observation, but the sum-
mary accounts of his observations are vague and obscure; Gal ton was a
mathematician, but he worked mostly upon material obtained second-hand
as the result of inquiry, so that it was not truly accurate. Mendel applied
himself to the study of the phenomena of heredity in garden plants; he se-
lected, to start with, certain easily observable characters — the colour of
the flowers, the shape of the seeds, the structure of the position of the blooms
— and he studied their modifications in different generations. He crossed peas
with white and red flowers; the hybrids then proved to be red throughout;
when, again, these hybrids were allowed to fertilize themselves, the succeed-
ing generations turned out to be coloured in a peculiar way: for every three
red individuals there was one white. These white, if self-fertilized, invariably
produced white offspring, one-third of the red remained similarly con-
MODERN BIOLOGY 591
stant, while the remaining red flowers repeated the above-mentioned colour
ratio.
Dominant and recessive characters
The explanation that Mendel gave of this strange phenomenon was as in-
genious as the observation itself; the fact that the flowers of the first hybrid
generation are red and acquire no intermediate colour between red and white
he accounts for by the red colour's being dominant over the white, which
latter character he calls recessive. In the succeeding generation both domi-
nant and recessive characters again appear; transitional characters cannot be
observed. From this he concluded that in the sexual cells, or gametes, there
exists no fusion of characters; in the hybrid red-white exist potentialities
for red and white side by side in the male and the female cells; when these
are united, the fusion must thus take place in accordance with one out of
four possibilities: red-red, red-white, white-red, white-white. This explains
the proportion between the offspring's qualities; the two single-coloured
combinations no longer vary, but remain constant, while those with the
double characters are capable of repeating the same four possibilities so long
as they exist. The same system of law-bound heredity was shown by all
the characters that Mendel investigated in various plant species. He after-
wards took up experiments with the crossing of bees, which, as is well
known, produce many different racial types, but, disappointed over the lack
of encouragement that he received as a result of his investigations into plants,
he did not publish the results of this subsequent research-work, and they
have now been lost to us. During his own lifetime Mendel's achievements
attracted no attention whatever; Nageli, as we have seen, found them ir-
reconcilable with his own theories, and the other botanists displayed utter
indifference. It was not until the turn of the century that Mendel's remark-
able results were rediscovered in connexion with the hybrid research-work
that was then being carried out. Three observers — de Vries, Correns, and
TscHERMAK — simultaneously pointed out the agreement between Mendel's
observations and their own results. Thenceforward Mendel's name has been
one of the best-known in biology; even among the general public his fame
has in more recent times competed with that of Darwin himself. Much sur-
prise has been expressed over the fact that Mendel's brilliant observations
did not attract greater attention, and the blame has been laid upon the un-
known journal in which they were published. One might with greater justi-
fication ask oneself whether any of the more important publications of the
time would have undertaken to print results of research so utterly at vari-
ance with the prevailing conception of biology. We have only to remember
that Mendel denies variability in those characters that he observed, whereas
all the biologists were just at the time seeking after variations as material in
5 92. THEHISTORYOFBIOLOGY
proof of natural selection; and then come these assertions as to absolutely con-
stant or constantly divisible characters from the pen of a monk in a monastery!
It would certainly have been a miracle if they had found support from the
generation that had been brought up on Haeckel's Natural History of Creation.
Universality of the Mendelian laws
Nevertheless, the Mendelian principle of cleavage now forms the basis of
all hybrid research. All characters that it has been possible to observe in
living beings have the quality of "mendelizing"; de Vries's statement that
mutations cannot be subject to this law has proved to be incorrect, and the
exceptions that have since been observed have actually been explained in
accordance with the same principle. Mendelian research has been carried on
to an ever-increasing extent year by year, both by theoretical observers and
by practical breeders. And Mendelism has stood the test in regard to the
improvement of seeds and domestic animals no less than in the theoretical
field; it is only by its aid that the practical improvement of breeds has been
successfully based on exact principles instead of on mere chance. In Scandi-
navia Mendelism has won many adherents; Johannsen has contributed
greatly to its advancement, and in Sweden H. Nilsson-Ehle especially has
applied it to practical purposes with universally acknowledged success; his
work for the improvement of seed cultures, carried on at Svallov, has been
done in accordance with its principles and has received widespread recogni-
tion. And both in Europe and America there are a great many Mendelian
students — to name only a few, W. Bateson and R. C. Punnett, in England,
Erwin Baur and Carl Correns, in Germany, the previously mentioned
A. Lang in Switzerland, L. Cuenot in France, T. H. Morgan in America.
As a result of the investigations of these and many others more and more
light has been throwm upon the whole complex and manifold profusion of
varied hereditary factors and their mutual relation. In this field, indeed,
there must be a vast amount of research material; it only remains to select
certain characters of importance in one respect or another and follow them
up. This has in fact been done, and numerous Mendelian students have taken
up various subjects for research at which they have worked with an ever-
increasing tendency to specialize. Of these subjects we shall examine one
somewhat closely — namely, that which has been made possible through
the application of the methods of modern cell-research to the problem of
heredity.
The Nlorgan school
The home of this cytological heredity-research has mainly been America.
The experimental biologist Edmund Beecher Wilson (born in 1856, pro-
fessor at Columbia University), to whom we have previously referred, had
already made valuable studies of the influence of the reproductive chromosomes
MODERN BIOLOGY 593
Upon heredity, but it was really a scientist who had been trained in the
same school — namely, the above-mentioned Thomas Hunt Morgan
(born 1866, likewise a professor at Columbia University) — who discovered
both the aim of this research work and the means for carrying it out, thereby
providing the study of heredity with a wealth of material by way of detailed
discoveries of far-reaching theoretical application, such as had never been
found elsewhere. His subject for investigation has been a small parasite
fruit-fly, Drosoplnla melanogaster, of which it has been said that it has appar-
ently been created by God solely as an object of heredity research. It repro-
duces itself with incredible rapidity and profusion — in heat an individual
requires only twelve days to develop from the egg to sexual maturity — it
is extraordinarily hardy and can stand every possible kind of experimental
treatment; its cells contain only four pairs of chromosomes, all of a differ-
ent size and easily recognizable; and, finally, it has given rise under labora-
tory conditions to a profusion of mutations, the factors of which have been
found to be constant and well adapted to Mendelian investigations. Morgan
and his numerous pupils have examined millions of these creatures, in the
course of which he has built up a methodology of his own and a terminology,
which, owing to its refined subtlety, is extremely difficult for anyone but a
specialist to comprehend. Among the results of this research work we note,
first of all, a number of fresh principles, as hard and fast as Mendel's origi-
nal ones. As had already been assumed, the hereditary factors are localized
in the chromosomes, and it has been discovered that the factors in the same
chromosome are not free, but invariably follow one another upon cleavage-,
they are "linked," as it is called. Thus, all the factors in this animal are
grouped into four linkage-systems, one for each pair of chromosomes. Fur-
ther, after a series of ingenious experiments it has been possible largely to
determine the position of the factors in each chromosome, and, finally, in
certain cases to establish the absence of parts of chromosomes or entire
chromosomes as being the cause of various external modifications — ■ that
is, a set-back for the theory of the absolute constancy of the chromosomes
in the same species. Besides this, it has to a large extent been made possible
to ascertain the part played by the previously mentioned sex-chromosome
in the determination of sex and in heredity — a problem which, as a matter
of fact, observers not belonging to Morgan's school have also studied. In
this province likely explanations have been found for a number of hitherto
incomprehensible phenomena of heredity; among those that are generally
known may be mentioned the inheritability of certain diseases, such as hae-
mophilia and colour-blindness in man, and, further, a number of cases of
heredity in the sphere of mental diseases. And, finally, mention should be
made of the valuable studies in regard to the relation of the chromosomes in
species hybrids; in this sphere may be mentioned, of Scandinavian students.
5 94 THE HISTORY OF BIOLOGY
H. O. G. Rosenberg, of Stockholm, who has investigated the hybrids
of Drosera, H. Federley, of Helsingfors, who established the fact that in
the butterfly hybrid the number of chromosomes is equal to the total of the
father's and mother's combined, and O. L. Mohr, of Oslo, who has carried
out independent investigations on the reproductive chromosomes and has
besides done valuable work in subjects dealt with by the Morgan school.
Heredity has been the most popular field of research of the age; it has
succeeded Darwinism in the way it has taken hold of the public mind and
has nowadays to serve as an explanation for anything that presents any diffi-
culty in the various spheres of life. Just as formerly it was natural selection,
so now it is the mixing of breeds that has to bear the blame for every kind
of circumstance and disparity even in human community life, in which po-
litical and social prejudices take good care that problems are at least not
treated impartially in the scientific sense. As a result of exact heredity-
research the theory of evolution has itself been directed along other lines;
phylogenetical speculations have for the most part been abandoned — at
least for the time being. Natural selection is certainly retained in principle
by some students of heredity — by Baur, for instance — but it is really of
no practical importance; the phenomenon cannot be observed and it is there-
fore not possible to fit it into a subject of research that is based on exact
observations. And while the old Darwinism operated with outward resem-
blance as a positive proof of common origin, heredity research has established
the fact that resemblance and affinity are not analogous terms, thus under-
mining the very foundations of phylogeny. Generally speaking, heredity re-
search goes to work in a more limited sphere; for the very reason that it
has become an exact science it has not been able to follow the old Darwinism
in its speculative ranging, but whatever may have been lost in the way of
the general conception of life has undoubtedly been won in the way of con-
centration on facts and reliable results.
There are still one or two other subjects for experimental research which
must be briefly dealt with in this chapter, and we shall now pass on to
these.
3. Biochemistry
Application of modern chemistry to biology
The science of chemistry as applied to biology has always afforded it valu-
able assistance in the search for an explanation of vital phenomena. In mod-
ern times, as is well known, chemistry has made splendid progress, and every
new step has at once had its influence on our knowledge of living organisms.
At the present day biochemistry is a line of research that, in point of the
THOMAS HENRY HUXLEY
From a portrait by the Hon. John Collier, reproduced in Some Aposfhs of
Physiology by William Stirling, Waterlow and Sons, Limited, London
OSCAR HERTWIG
MODERN BIOLOGY 595
value of its results, competes well with experimental morphology and hered-
ity research. Unfortunately, these results are far less accessible for the pur-
poses of popular presentation than any other of the advanced spheres of
biology; in fact, biochemistry requires a very special technical training for
both its students and its critics. Nevertheless, space may be found here for
a few brief indications as to the most important progress that has been made
in its various special provinces in order to complete the picture of the gen-
eral progress made by experimental biology in modern times.
That branch of chemical research which has had the greatest success
in our own day and has excited the keenest interest is undoubtedly physical
chemistry; each stage of its progress has at once been applicable to the liv-
ing substance. Biology has actually led the way in certain physico-chemical
discoveries, as in the question of osmotic pressure, in which the results of
Pfeffer's and de Vries's research-work formed the foundations on which
scientists have subsequently built further. On the other hand, the modern
theory of solutions, as created, among others, by van t'Hoff, Arrhenius,
and Nernst, has contributed towards explaining a great many biological
phenomena; as an instance may be mentioned the above-described partheno-
genetical phenomena in eggs that have been subjected to hypertonic saline
solutions; further may be quoted the part played by hydrogen ions in the
metabolism of the fluids of the body: they play a decisive part especially in
producing respiratory irritation, while other ion-combinations have been
found to be necessary for growth in individuals and organs.
Colloid chemistry
Of special significance for forming a conception of the nature of protoplasm
has been modern colloid chemistry; it has founded a province of its own,
with its own methods, which have made it possible to study far more closely
than before the most minute structural details in the category of elements
in which living substance is included. Thanks to these accurate observations
and experiments, the granular and vacuolized structure of plasm has been
given a far more natural explanation than that once given by Biitschli in
his froth theory. It has in many instances been possible to compare the mu-
tual interpenetration of the various structures even with physico-chemical
metabolistic phenomena occurring in inanimate colloid substance, while,
on the other hand, the old dispute as to the solid or fluid nature of plasm
has lost all point; the intrinsic character of and changes in the colloids are
investigated on entirely different principles and have given rise to problems
utterly different from this old question of the state of aggregation. Instead,
extensive investigations have been made into the question of the penetra-
bility of cells and tissues by solutions of various kinds; in this sphere
especially Charles Ernest Overton (born in England in 1865, and after
studying in Germany appointed professor in Sweden) propounded a theory,
5 96 THE HISTORY OF BIOLOGY
which has been both contradicted and supported by others, that the cellular
membranes consist of a peculiar category of elements called lipoi-ds^ elements
dissolved in these permeate the cell-v/alls, and vice versa.
Ferment chefnistry
With colloid chemistry the modern chemistry of ferments is closely allied.
In regard to ferments it has been known for a hundred years that through
their presence in extremely small quantities they are capable of causing vari-
ous chemical changes in large masses of the substances on which they are
acting. In modern times this influence they exert has been compared with
the catalytic effect of certain inorganic substances (first pointed out by Ber-
zelius), as, for instance, the part played by acid in the production of ether
out of alcohol. Of fundamental importance is the discovery that not only
are phenomena of disintegration induced by ferments, but also synthetic
processes, as, for example, the production of starch in the leaves of plants
through the chlorophyll granules under the influence of sunlight. The fer-
ment syntheses of fats, albuminous substances, and other products of vege-
table and animal life, which have been the objects of special study, are in
themselves of immense interest, but they cannot be discussed in detail here;
nor can we describe the extremely subtle investigations that have been carried
out in connexion with the co-operation of various ferments within the same
vital unit.
In connexion with fermentation research mention should be made of
the process of internal secretion, our knowledge of which has increased more
and more in recent times. Bernard, whom we mentioned earlier, may claim
to have been its discoverer. As we have seen, it was he who established the
fact that the liver produces substances which are directly carried away by
the blood. Charles Edouard Brown-Sequard (1817-94) succeeded to Ber-
nard's professorship after the latter's death. The son of an American father
and a French mother, Brown-Sequard studied at Paris and afterwards worked
in England and America, but later on returned to France. He became fa-
mous for his valuable investigations in the sphere of neuro-physiology, but
more especially for his experiments, which he published in his old age, with
the injection of extract of genital glands, whereby he demonstrated that
these glands contain a special secretion inducing sexual desire. The experi-
ments in themselves were somewhat clumsy and were published with a good
deal of advertisement, especially as regards the power of rejuvenating the
individual, which was promised as a result of them; all the same, it cannot
be denied that through them attention was drawn to a fact that has since
been investigated by others with greater thoroughness than before. The in-
terstitial gland of the genital organs has been anatomically investigated by
the two collaborators Ancel and Bouin, of Nancy; its physiological func-
tion and, in general, the influence of the genital organs upon the vital
MODERN BIOLOGY 597
manifestations of the body have been studied, inter alia, by J. Meisenheimer,
ofLeipzig,whoexperimentedonthelarvie of butterflies, and by Eugen Stein-
ACH, of Vienna, who studied vertebrates from the same point of view. Stein-
ach has succeeded by means of operations in influencing the sexual character;
the genital glands of male and female animals have been interchanged, with
the result that both the outward appearance and the sexual behaviour of the
animals have been correspondingly altered. Otherwise, Steinach is best known
for his having resumed rejuvenation experiments similar to those of Brown-
Sequard, but more carefully carried out both in theory and in their details.
These experiments have won him a fame that, owing to the nature of the
problem, has become associated with much of the glamour that surrounded
his predecessor; moreover, they appear already to have exhibited defects that
have rendered them impossible of realization in practice.
Internal secretion and rejuvenation experiments
In the mean time our knowledge of internal secretion in other spheres has
increased with great rapidity. So-called endocrine glands have been discov-
ered in large numbers; among them may be mentioned the suprarenal cap-
sules, hypophysis, the thymus and the thyroid gland; and our knowledge
of their functions has at the same time been extended. But other organs have
also become known as producers of internal secretions, or hormones, as they
are also called; the small intestine produces such a secretion, which is termed
"secretin" and which, when conveyed through the blood to the pancreas,
induces secretion in that gland; another similar substance is produced in con-
nexion with the pregnancy of female animals and causes the segregation of
milk, and a third is the product of a special cell-category in the pancreas
and has a definite influence upon the metabolism of the body, in that its
absence induces diabetes. On the whole, many of these internal secretions
have become known only by indirect means, through the diseases that arise
if the organs which produce them are injured or removed.
Serology
In modern times serology forms a separate field of research, embracing one
of the most important chapters in practical medicine. Pasteur should be
named as its founder, while later Behring, Ehrlich, and their pupils have
particularly distinguished themselves in that subject. It has been established
that the danger of the disease-producing bacteria lies in the fact that in the
course of their multiplication in the body they produce special isolatable
chemical compounds having a specific poisonous effect, which have been
given the name of toxins; the body reacts against these by forming similarly
specific elements, antitoxins, which counteract them. To isolate these latter
and to use them as a counter-poison has in many cases proved the only w'ay
for medical science to cure infectious diseases. By injecting the bacterial poi-
son into experimental animals, these latter have been allowed to produce
598 THE HISTORY OF BIOLOGY
the antitoxin, their blood-serum afterwards being used as a counter-poison
against the disease. The actual elements — both toxins and antitoxins —
that operate in this process always exist in very small quantities, and for
that reason and owing to their complicated composition it has not been
possible to analyse them; their quantitative effect upon one another has
nevertheless been investigated by Svante Arrhenius and others.
In connexion with this research work the nature of the blood in gen-
eral has been the object of very detailed investigations, often with quite
remarkable success. Among the best-known of these is the precipitin re-
action, which was discovered by Paul Uhlenhuth (born 1870, Behring's
successor at Marburg). If we take serum from an animal — for instance, a
dog — and inject it into a rabbit, we obtain from the rabbit's blood within
some days a serum that produces precipitation in the serum of the dog's
blood. This reaction is specific and can therefore be utilized in medico-legal
investigations in order to distinguish, for instance, human blood from ani-
mal blood. In this, however, similar forms act in an identical way; for ex-
ample, human blood and the blood of anthropoid apes produce the same
reaction. When this chemical resemblance between allied organisms was dis-
covered, it excited great enthusiasm as a phylogenetical argument; closer
consideration, however, at once makes it clear that this agreement of chemi-
cal composition demonstrates just as much or just as little as the morphologi-
cal resemblance that can be demonstrated by the old comparative method;
the fact that resemblance in bodily structure, food, and habit is accompanied
by corresponding chemical agreement is essentially so obvious that the con-
trary would be more surprising.
We may here cite one more example of discoveries in this sphere. We
have previously mentioned how the Russian naturalist Ilja METscHNiKorF
(1845-1916), who after studying at German universities was a professor at
the Pasteur Institute in Paris, produced the so-called phagocyte theory; he
found, as indeed Haeckel had already observed, that the white blood-cor-
puscles absorb foreign substances into the body; in particular, he discovered
that the leucocytes in this way free the body from bacteria that enter it, pro-
vided the latter are not too strong and do not get the upper hand. In more
recent times it has been found that special substances are produced, which
have been called "opsonins," which possess the ability to increase the
leucocytes' power of killing the bacteria. These substances, however, are
at present little known, but they are, of course, of the very greatest interest.
They have been mentioned here as examples of the wide possibilities with
which modern biochemistry has to reckon. The immense practical benefits
that this research work has brought humanity can only be hinted at here;
some of the theoretical speculations to which it has given rise will be dealt
with in the following.
MODERN BIOLOGY 599
V.
4. Animal Psychology
Animal psychology would rightly have been made part of experimental
biology if it had consistently remained on the basis of empirical research.
This, however, has been far from being the case; on the contrary, ideas
about the psychic life of the animals have varied infinitely up to the pres-
ent day, from the standpoint of primitive man, who ascribes to the animals
an intelligence of the same kind as his own, to Descartes's conception of
animals as completely automatically operating mechanisms. The reason for
this confusion is, of course, to be sought in the vague ideas of human psy-
chology, which have varied according to the general point of view taken
by the different schools of philosophy. It was not until the middle of last
century that an exact psychological research began to appear, thanks to
precursors like Theodor Fechner, Wilhelm Wundt, and their pupils. This
empirical psychology treats of psychical phenomena just like any other ma-
terial for observation; it is, as Hoffding says, "no more bound to begin with
an explanation of what the soul is than physics is bound to begin with an
explanation of what matter is." Unfortunately, those biologists who have
dealt with the phenomena of animal psychology have by no means always
taken this warning to heart; not infrequently they have followed Haeckel's
bad habit of involving themselves in speculations upon the soul as such
without having the qualifications to give critical treatment to this com-
plicated problem.
Self-observation the joiindation of psychology
The foundation of all empirical psychology is self-observation; on this point
an animal psychologist of the old school, such as Romanes, is in agreement
with a modern experimental psychologist of the type of Alfred Lehmann
(1858-19x1), a pupil of Wundt, professor at Copenhagen. Lehmann main-
tains that it is only in one's own consciousness that one can observe psychical
states and functional manifestations; the assumption of psychical phenomena
in other creatures depends on whether these latter are seen to act in given
circumstances as man would do in the same circumstances. As regards one's
fellow human beings, this conclusion can be confirmed by means of language,
but such a mode of control is wanting in animals; as far as the vertebrate
animals are concerned a good deal can be concluded from their bodily struc-
ture where it agrees with that of man, but even this check is lacking in the
invertebrates. As a general principle of animal psychology there remains,
then, according to Lehmann, the principle of ascertaining by experiment
whether the animal can adapt itself to new and unexpected situations;
whether it can by learning from experience modify its actions to suit the
conditions. If this is done, then we have the right to assume the existence
of an individual psychic life in the animal — to assume life-phenomena of
6oO THE HISTORY OF BIOLOGY
a different kind from the instinctive adaptation to normal conditions of life
which characterizes all animal life and which is based upon nervous reflexes
or simply upon tropisms induced by chemical and physical reactions, such
as are observed even in the most primitive of organisms. Such experiments
with individual experiences as their object are extremely difficult to carry
out, however, and still more difficult to interpret aright. On the one hand,
animals should be brought into situations which prove that they can learn
something new, but, on the other hand, they should not be faced with situa-
tions which involve violence to their true nature. The road to a true insight
into the subject is thus both long and difficult, and this explains to some
extent why so many contradictory statements and theories have arisen in
this sphere, even among observers who have been comparatively successful
in freeing themselves from preconceived opinions. The confusion has, of
course, been increased by so many students and dilettanti, for " Darwinistic "
purposes, ivanting necessarily to find in the animals as great and as human-
like an intelligence as possible. On the other hand, there have not been
wanting in modern times zoologists who have seen in animals nothing but
reflex mechanisms, and that, too, not merely as a result of their holding a
conservative view of life, but quite as often owing to ultra-radical views —
through endeavouring to restrict as far as possible the part played by the
psychic life in nature. We shall here cite in brief a few examples of different
views on these problems.
Insect psychology
One subject that has specially interested animal-psychologists from ancient
times has been the psychology of the insects, particularly of the community-
forming Hymenoptera. Here in earlier times the imagination and credulity
have combined to celebrate veritable orgies of the fancy; one reads with un-
feigned amazement of all that people even in the latter half of last century'
imagined they could observe in the ant-communities. A much more sober
atmosphere has latterly prevailed, and it has now begun to be realized that
most of the actions of the ants must after all be due to inherited instinct.
A very prominent observer in this sphere is the Jesuit father Erich Wasmann
(born 1859), who has especially elucidated a number of facts in regard to
the ants' relations to many different kinds of parasites, which swarm in
the ant-heaps and which are often carefully looked after. Wasmann has
otherwise appeared as a keen opponent of Haeckelian monism and has elab-
orated in opposition to it a history of creation approved by Roman Catholic
authorities, in which Darwinism in most of its aspects has been ingeniously
2 As an instance may be cited a story related by many authors — Romanes, for example
— of an American species of ant, according to which those ants whose duty it is to guard the
communities do so formed up in a regular square; within the square the workers carry out a
great manv equally intelligent and well-ordered movements.
MODERN BIOLOGY 6oi
introduced, though at the same time the old doctrine of creation has been
retained. In regard to the psychic life of the animals, its intrinsic resemblance
to human life is, of course, denied, and accordingly the reflexive and in-
stinctive functions are sharply emphasized and the idea of individual con-
sciousness rejected. Nevertheless, even the insects are allowed a certain power
of methodical action. Albrecht Bethe comes to a far more radical conclu-
sion, based on entirely opposite grounds; by means of extensive and in part
highly ingenious experiments he endeavours to prove, and in many cases
actually succeeds in doing so, that the actions of the ants are pure reflex-
actions. On the other hand, the famous student of psychiatry and psychology
AuGusTE FoREL, of Geneva, holds another view; after a series of careful ex-
periments he believes that he has been able to prove that insects can actually
learn by experience and thus possess intelligence. Another distinguished ob-
server of insect life who arrived at partially similar results was the well-
known English naturalist, banker, and philanthropist John Lubbock,
Lord Avebury (1834-1913). A very important animal psychologist was
George John Romanes (1848-94}, professor at Oxford, an enthusiastic sup-
porter of Darwin, whose theory he defended in a number of writings, in
which he especially attacked Weismann for his denial of the heredity of
acquired characters. In particular, he carried out experimental investigations
into the psychic life of the higher animals, which he believed to be — in
kind, if not also in degree — similar to that of human beings. Like Darwin,
however, he has accepted without criticism stories and anecdotes derived
from foreign sources, but his own observations are very keen. Animal psy-
chology has been dealt with experimentally with special keenness in Amer-
ica; among its pioneers in that country may be mentioned R. M. Yerks,
who has carried out a long series of experiments in order to try to discover
the power of animals to learn by experience; he is especially well known for
his "labyrinth," in which he placed animals, who then had to find their
way out as best they could.
As a general result of the work of these investigators it may be men-
tioned that the vertebrate animals, at least the higher, can certainly acquire
knowledge from experience; whether the higher invertebrates can also do
so would seem to be more doubtful. But even experience can be exhibited
in different ways; either the animal finds its way in every fresh case to a new
experience, independent of what it has gone through before, or else it really
possesses the power to retain its experiences in the memory and to profit
by them in new situations. The former power is, of course, the more primi-
tive, as it is also the more usual in the animal kingdom; the latter, the ra-
tional power of adaptation, has certainly been observed to exist only in a
few of the very highly developed animals. Still more debatable are the cases
in which the power of grasping relations of number and other abstract ideas
6oi THE HISTORY OF BIOLOGY
has come into question. Isolated cases have been known of old of alleged
purely human intelligence in domestic animals. Shakspere in one of his com-
edies alludes to a horse, on show at the time, that was able to count; in
our own day one or two similar cases have certainly given rise to much dis-
cussion. Even professional biologists have believed, after investigation, in
the famous horses of Elberfeld, which performed operations of counting in
higher mathematics and other equally remarkable feats. On the other hand,
the authenticity of these cases has been very keenly disputed — Alfred Leh-
mann describes them unreservedly as self-deception on the part of the spec-
tators or conscious duplicity on the part of the exhibitor. At best, however,
it could only be a matter of abnormal receptivity to training, which cannot
be considered in any respect to characterize normal intelligence in the ani-
mals and thus cannot offer any guidance for judging the development of
the intelligence in the animal kingdom, which, however, has been very
generally asserted by those who have felt convinced of the reality of the
feats exhibited. In cases like these keen criticism is more than ever essential;
unfortunately many even distinguished biologists have shown themselves,
owing to their preconceived ideas, far stronger in their beliefs than clear
in their judgments, and this has been not least apparent in that much dis-
puted chapter, the psychic life of the animals.
CHAPTER XVIII
MODERN THEORETICAL SPECULATIONS
I. Mechanism and Vitalism
IT WILL HAVE BEEN SEEN from the previous chapter that biology in our
own day has distinguished itself far more through practical detailed re-
search than through theoretical speculations. Actually no further gen-
erally accepted theory of life such as that offered by Darwinism has been
discovered; instead, there has prevailed, a restless search for fresh grounds
on which a theory of life might be built up. Many have been the roads along
which the search for the theoretical solution of the problem of life has been
made since the turn of the century, and they have run in widely differing
directions. On the one hand, we find repetitions in a newer form of the old
materialistic and mechanistic theories of life from the days of Vogt and
Haeckel; on the other hand, vitalistic ideas that have looked for support
in Bichat, in Stahl, in Aristotle. The old antagonism, Christian conservatism
versus Darwinistic radicalism, which half a century ago resulted in the for-
mation of parties, has now been essentially adjusted, though in no wise
everywhere eradicated; in this respect it must be acknowledged that the
struggle in modern times is more definitely a matter of facts than it was
previously, at least among students, but the differences of opinion have in
many quarters certainly been sharp enough. It is possible to give here only
a few examples of views taken from the rival camps, this as a final summary
of the position of biology as it stands in the present generation; unfortunately
it is more difficult than ever to draw conclusions from them as to the direc-
tion likely to be taken by evolution in the future.
Max Verworn (1861-19x1) was born in Berlin, studied there and at
Jena, and eventually became professor of physiology at Gottingen. He was
a pupil of Haeckel and throughout his life retained both his admiration for
his master and his association with Haeckel's fundamental ideas. To him
the biogenetical principle as well as the theory of selection always was a
proved fact; of the more recent contributions to the hypothesis of evolution
he embraced de Vries's mutation theory, but he showed no interest in Men-
delism. On these foundations, however, he built up a life theory of his own
603
6o4 THE HISTORY OF BIOLOGY
that has found lively support in many quarters. He calls it "cellular physi-
ology" and bases it, as its name implies, on the doctrine of the cell as the
unit of life, which he maintains repeatedly, and with greater emphasis than
anyone else has done, against all theories of smaller, independent entities,
such as Altmann's granula. To him the body is exclusively a cell-state; the
formation and co-operation of the organs have little interest for him and are
dealt with only in passing; Roux's name is mentioned as little as Mendel's
even in the most recent edition of his Allgememe Physiologie. He chiefly oc-
cupies himself with the living substance and its nature. And though he has
removed from this field of research those hypothetical life-units of a mechan-
ical kind that played such a large part in the theories of his predecessors
and contemporaries, he has nevertheless reverted to the same unworkable
imaginative system of thought, although on different grounds; that is to
say, he has imagined a chemical unit of life of an albuminoid character,
which he terms "biogen," whose chief characteristic apparently consists
in an extreme chemical lability: a constant and simultaneous alternation of
disintegration and reconstruction. In this chemical change consists, accord-
ing to Verworn, the true essence of life; there is no difference between ani-
mate and inanimate except that which is brought about by the extraordinary
metabolistic possibilities of the biogen molecule. Verworn, it is true, admits
that the biogen is really as hypothetical as plastidules or micellx ever were,
but he has the same weakness as other speculative biologists for allowing
hypothesis, when it is once completed, to stand for fact. It is a more serious
matter, however, that the biogen cannot be reconciled with modern biochem-
istry; the latter' s representatives have fairly unanimously condemned the
whole hypothesis, maintaining that there exists no other albumen molecule
than that which is already known to general chemistry. "There is no such
thing as dead and living albumen any more than there is dead and living
sugar or fat, and the reactional powers of the protoplasm depend upon the
co-operation of its various component parts in definite proportions" (Hober).
Verworn, however, paid no heed to these objections, especially as modern
biochemistry did not interest him in the least; he makes no mention of the
progress of colloid chemistry, but instead discusses the old problem as to
whether plasm is solid or fluid.
Verworn s conditionism
On the whole, then, Verworn went his own way, heedless of the progress
of contemporary science; he had made it a principle, one of his biographers
relates, not to read too much, but to think for himself. He was, moreover,
of an ardent disposition; ethical and social questions interested him keenly,
and every new-year's eve he ceremoniously inaugurated the coming year's
work by sitting down at his writing-desk at the stroke of twelve. Having
such a temperament he was naturally inclined to indulge in philosophical
MODERN BIOLOGY 605
Speculations upon the main problems of life; he elaborated a theory of his
own that he called "conditionism," which was to replace the old ideas of
cause and effect by a conception of all phenomena's being due to a multi-
plicity of simultaneous conditions; what he was trying to get at was, of
course, the old contrast of things and phenomena and thereby ultimately
the contrast between the physical and the psychical, which he would replace
by an all-comprehensive "psycho-monism." But Verworn was no clear-
sighted thinker; conditionism is reminiscent of Mach's phenomenalism,
which, however, is far better thought out, and psycho-monism merely leads
to a great many far-fetched and unnatural attempts to get away from the
actually existing and observed difference between animate and inanimate, in
which the hypothetical "living" albumen must play the part of a universal
remedy for all difficulties. Through his enthusiasm, his brilliant style, and
his undeniably valuable contributions to the problem of the cell as a physi-
ological unit of life, Verworn has at any rate been an important personality
among the biologists of his age.
Loeh on the movements of animals
A MECHANISTIC explanation of nature on entirely different grounds has been
produced by the experimentalist Loeb, who has been mentioned in the pre-
vious chapter. He was, as already pointed out, a pupil of Sachs, whose
studies of the tropisms of plants became the basis of his entire conception
of life. In his earlier years he himself carried out a number of valuable in-
vestigations on the subject of tropisms in the lower animals; since then he
has made the above-described experiments on parthenogenesis in eggs caused
by chemical reagents. The general theory of life that he set up is actually
based on these two classes of experiments. He regards, as far as is possible,
the movements of animals as tropisms caused by external influence; when an
animal moves towards the light, there actually takes place through the effect
of the light an oxidization of certain elements in the animal, and this causes
the movement; other movements, again, are induced by chemical associa-
tions that arise directly in the innermost being of the animal, as, for instance,
in the mating-flight of insects. On these facts he bases a "mechanistic con-
ception of life," which, however, he hardly succeeds in formulating in a
very convincing way. Indeed, he has no idea of a scientific student's duty
of first thinking out his theories; when his theory suits one case, it is at
once made to hold good for all cases without further investigation, and if
it does not do so, then the inexpedient cases are simply passed over. He gives
an account, for instance, of the phototropism of the Aphida:, on which he
carried out most ingenious experiments, and he traces the phenomenon to
the said oxidizing process — the fact that this phenomenon upon repeti-
tion proceeds with greater rapidity "may be brought about by" the lactic
acid produced by the muscles upon movement. Thus the phenomenon in
6o6 THE HISTORY OF BIOLOGY
question is ascribed to a cause that is stated to possess a general significance
for all vital phenomena, but when shortly afterwards it is declared that the
worker-ants do not exhibit any such tropism, no attempt is made to explain
this notable exception. And the same lack of consecutive thought is dis-
played everywhere. The development of the egg is explained as being due
to oxidization; that this development can be caused by such diverse external
influences as, on the one hand, a solution of acid and, on the other, the
prick of a needle certainly evokes some surprise, but no more than that the
whole phenomenon is after all accounted for as being a physico-chemical
process, it not being considered at all necessary to discuss the most remark-
able feature of all — namely, the nature of the egg itself. Indeed, in the
opinion of Loeb, there exist no structural conditions whatever; there is
hardly any question of the organism's possessing a chemical composition
of its own; all that takes place in the organism is the result of outside im-
pulses, the result being that no discrimination whatever is made between
one life-phenomenon and another, whether it is a question of sea-urchins,
insects, or frogs. The goal to be attained is, as we have said, a mechanical
explanation of life, but just because of this exclusive interest for external
influences the explanation proves to be essentially negative — a denial of
the existence of any operating forces other than the said external influences.
When he comes to discuss more complicated problems, Loeb shows the most
amazing lack of criticism; he gives an account of Mendelism and declares
that the riddle of heredity is thereby solved, but a little further on he as-
serts the exact opposite, that any ossean can be crossed with any other os-
ean;^ he himself has kept such hybrids alive for a month. The explanation
of what has taken place is manifestly a parthenogenetical development of
the eggs through the "activating" influence of foreign sperm; but Loeb
literally declares that it is possible to obtain a hybrid between a salmon and
a flounder. With such facility in drawing conclusions and such irresponsi-
bility as to their consequences there need be no limit to one's flights of
imagination, and, moreover, one is free in the long run to dispense with
all exchanges of views with scientists possessing a normal sense of respon-
sibility. Loeb is without doubt a brilliant experimentalist, and as such he
deserves mention among the pioneers, though among biological thinkers he
can claim no place.
Vitalistic explanations of life
On the whole, the mechanistic speculations in the sphere of modern biology
give a somewhat monotonous impression and it is therefore hardly worth
while making the acquaintance of any more representatives of this line of
thought. Those who are constantly making the assertion that there is no
1 See The Mechanistic Conception of Life, p. 2.4: "It is possible to cross practically any marine
teleost with any other."
MODERN BIOLOGY 607
essential difference between animate and inanimate very quickly lose ail ap-
preciation of what is truly characteristic in living matter and its metabolistic
phenomena, which must otherwise be the chief interest even of those bi-
ologists who maintain the old assertion, already proclaimed by Kant, that
only material phenomena can be subjects for natural-scientific treatment. It
is not, then, the idea — in itself justifiable — of limiting discussion to the
chemical and physical manifestations of the phenomena of life that consti-
tutes the weakness of these mechanistic theories of life, but the stubborn
insistence upon the rough comparisons between phenomena in animate and
inanimate nature — comparisons, in fact, the weakness of which would un-
doubtedly be realized by their proponents if the latter were not really trying
all the time to lay the foundations of some kind of general philosophical
theory extending far beyond the bounds of natural science. When Verworn
discusses and denies the possibility of the immortality of the soul, he is
arguing, from the natural-scientific point of view, about nothing at all, for
though, as we have seen, biology studies psychical phenomena, it does not
imply that it has anything to do with the impossible problem of what the
soul is or is not. And if we ask, quite apart from such metaphysical quibbles,
whether all observable material processes in the living organism or its parts
can be directly derived from known material processes in inanimate nature,
the answer even today must still be in the negative; those who have at-
tempted to do so have either reverted to gross schematism or else drawn a
bill on the possible progress of tomorrow — an unworthy manner of wrig-
gling out of the fact of the problem's insolubility. It may at once be assumed
that the future will bring us nearer the heart of the problem; whether it
will ever be entirely solved we know no more than the truth, laid down by
Herbert Spencer and many others, that the capacity for knowledge is limited
and the most general laws in existence must therefore remain unexplained.
Strictly speaking, the same causes have brought about the popularity
of the mechanistic theories of life as those that at one time produced so
many editions of Haeckel's Natural History oj Creation. It is obvious, however,
that a reaction against this conception of life was bound to set in, owing to
the disappointment felt over its splendid but unfulfilled promises. And so we
find, even before the turn of the century, vitalistic theories of life of various
kinds being produced, supported by representatives of no small importance,
as regards both their numbers and their attainments. And during the pres-
ent century their number has still further increased; true, they can nowhere
be said to have dominated the situation, but the part they have played has
been quite an important one, many of them having had a perceptible in-
fluence even in circles in which vitalistic or spiritualistic ideas have other-
wise never been very highly appreciated. They have for the most part come
from the physiologists, while the morphologists, v/ith far fewer exceptions.
6o8 THE HISTORY OF BIOLOGY
have adhered to the standpoint that they had maintained since the appear-
ance of the descent theory. The views of some of the vitalists have arisen
in connexion with certain religious or social principles, as in the case of
the above-mentioned Jesuit priest Wasmann, who, owing to his ecclesias-
tical point of view, was bound to feel attracted to a theory of life that offers
the possibility of a spiritualistic explanation of existence. So, too, the Luth-
eran conservative botanist and politician J. Reinke, mentioned in the pre-
ceding chapter as an opponent of Haeckel and known for his violent polemics
against materialism. Greater impartiality has been shown by the physiolo-
gists GusTAv BuNGE and R. Neumeister, both of whom have interested
themselves in the question of whether the phenomena of life can originate
merely in physico-chemical causes. Bunge sustains his arguments especially
upon the complicated structure and vital manifestations of the cell, which
cannot be explained on a physical or chemical basis; moreover, he takes as
his premisses that life as a whole cannot be studied except through self-
observation and is therefore a psychical process. Neumeister, on the other
hand, maintains against Haeckel and his supporters that the origin of life
is a transcendent problem and that the psychical phenomena cannot be de-
rived from the material. This criticism of contemporary materialism is in
itself certainly justified, but it actually implies a negative attitude; to try
to substitute for it an imaginary life-force is only to create a fresh compli-
cation of the problem; it militates against our striving after simplicity, as
Henle once said. The most pronounced vitalists of our own day have been
fully conscious of the fact, but they have deliberately exceeded the bounds
of exact research and gone over into the sphere of abstract speculation.
The ' ' autonomy of life-phenomena
The most interesting of these modern philosopher-scientists is undoubtedly
the previously mentioned experimental biologist Hans Driesch, inasmuch
as his history shows the natural development of the consistent vitalist from
biologist to metaphysician. Born in 1867, the son of a wealthy merchant,
he was allowed full liberty, regardless of prevalent trends of thought, to
devote himself to science. His start as an experimental biologist has already
been described above, as also how he interpreted his experiments in a mark-
edly epigenetical way, thereby finding himself opposed to Roux and his pre-
formation theory. Markedly antagonistic was also his attitude towards
Darwinism, which when still a youth he declared to be a thing of the past.
Owing to these two facts — the results of his experiments, and his anti-
Darwinism — he adopted from the outset an aggressive attitude towards
the older biological school, which accounts for his keen insight into its
weaknesses. Moreover, his is a pronounced speculative nature, with wide
philosophical interests and corresponding erudition, and he has been very
deeply attracted by the Aristotelean abstract construction of life-phenomena.
MODERN BIOLOGY 609
On these grounds he has produced his arguments in proof of "the autonomy
of life-phenomena"; he asserts that a living being forms a "harmoniously
equipotential system," by which he means that a hydra or other primitive
organism can be regenerated out of severed parts; this proves, to his mind,
that the animal is not a machine, for a machine cannot be evolved out of
its parts. This antithesis — machine and living being — he is constantly
bringing forward as a proof of the impossibility of deriving the animate
from the inanimate, and by drawing a comparison between them he comes
to the same negative result as the vitalists referred to above, though on
markedly abstract and schematic grounds.
The ' ' entelechy
He is not content with this, however, but seeks to discover what life really
is. The answer is summed up in an expression borrowed from Aristotle: "an
entelechy." By this word Aristotle meant the potentiality which is inherent
in matter and which achieves reality to the extent of matter's development
into an ever higher and higher form (see Part I, p. 36). Driesch has likewise
a marked interest for form in living nature, which, he considers, lends itself
far more readily to "philosophical analysis" than metabolism does. By en-
telechy, however, Driesch means something far more involved : it is supposed
to mean "something that carries its purpose within itself." It is thus the
functional adaptation of living beings that is here indicated, but in entering
into a profound and far-reaching analysis of the idea Driesch becomes in-
volved in a maze of abstract speculations, which become still more difficult
to understand owing to the extremely complicated terminology he employs;
really we have to go back to the heyday of Hegelian philosophy to find the
counterpart, in point of difficulty of comprehension, of Driesch's definitions
and characterization of the phenomena of life. This much, however, can be
gathered from them, that as the ultimate proof of his vitalism he cites his
own personal consciousness; it is thus, apparently, that we are to interpret
his expression " phenomenological idealism," which, according to his con-
ception, leads directly to vitalism, at any rate as far as his own body is
concerned. He then draws the same conclusion in other living bodies. But
this is certainly, if anything, pure metaphysics; it has nothing to do with
biology; to give a detailed account of how the idea of the relation of en-
telechy to matter is further developed would in such circumstances be super-
fluous, all the more so as here the incomprehensibility of his language ex-
ceeds all bounds; sentences such as the chapter heading: " Entelecbie bex,ieht
sich auj den Kaum und gehort daher %}ir Natur, aber Entelecbie ist nicht im Kaum,"
which is afterwards explained as follows: " Sie wirkt nicbt hn Kaum, sie wirkt
in den Kaum hinein,'' do not make the reader much the wiser. The same is
true of such a statement as that " Mater ie ist nicbt einmal in irgend einem
Sinne die Grundlage des Lebens." The new formula that Driesch gives in this
6lO THE HISTORY OF BIOLOGY
connexion to the law of the conservation of energy is a matter for the physi-
cists to consider. Driesch having thus already at an early stage taken up a
position essentially on the other side of the boundary line between meta-
physics and empirical research, he has ultimately adopted this step formally
as well; he is now professor of philosophy at Leipzig and in that capacity
has been engaged in speculations of the most abstract kind.
Another vitalist of whom a good deal has been heard is Emanuel Radl.
He was born in Bohemia in 1873, studied at Prague, and has been a lecturer
in physiology there. His research work has concentrated partly on physio-
logical subjects — he has dealt with the tropisms in the lower animals —
and partly on the morphology of the brain. He is best known, however,
for his Geschichte der biologischen Theorien der Neu^eif, a much read and widely
quoted work, the first part of which has come out in two editions, which
are essentially different from each other. To examine this work properly,
however, we must have some knowledge of its author's biological stand-
point, which is clearly apparent in his most important monograph, Neue
Lehre vom xentraUn Nervensysfem. In its introduction the author shows that
he holds particularly broad views, and, on the other hand, that he is fully
convinced of the soundness and future value of his own ideas. The Darwin-
istic morphology is rejected as a soulless description and specification of dif-
ferent developmental forms one after another; he gives slightly more credit
to experimental evolutional physiology on the lines of Roux, but the science
that in Radl's view has the greatest future is that of ideal morphology, of
which he himself is an exponent. As a pioneer of this science he mentions
Geoffroy Saint-Hilaire, and also, though in a less degree, Cuvier; its aim
is stated to be to discover the ideas according to which the forms of living
organisms are constructed. "Many ideas compete at the root of organic life
for precedence, and an ideal structure forms the basis of every organism."
This must be sought for by means of comparisons throughout the animal
kingdom, for only thus can we gain any knowledge of the fundamental ideas
of existence. This is, of course, simply the idealistic morphology of the begin-
ning of the nineteenth century over again. When it comes to working out his
idea in detail, however, we find only a collection of disjointed sentences taken
from other authors, in conjunction with his own, not always very convincing,
observations concerning the object of his investigation, the central nervous
system. Apathy's fibrilla theory is thus maintained as against the doctrine
of neurones; Radl himself describes in word and illustration one category
of fibrillar, existing, according to him, throughout the animal kingdom,
which are convoluted in a special way at the entrance to every ganglion
and which he terms "cascade fibrillar." They do not, however, give the
impression of being very natural and would appear rather to have arisen
through being cut obliquely or through the ordinary nerve-fibres' being
MODERN BIOLOGY 6ll
wrongly fixed. The formal points of agreement in the nervous system, upon
which ideal morphology is based, must, on the whole, seem rather unin-
teresting to a morphologist of the old school, for they prove nothing new;
but Radl imagines that his method has achieved extraordinary results. The
theory of organic structure is to prove capable of building up a magnificent
philosophy of its own, "such as speaks to us out of the Pythagoreans' theory
of harmony, out of Plato's doctrine of ideas, and out of the romantic rev-
eries of the obscure German natural philosophy." Radl's conception of the
world as a " Schopjung des sie betrachtenden Geistes" is undeniably reminiscent
of the latter; in fact, his expressed intention, underlying the entire work,
is to excite surprise, "for it is through being surprised that man has now and
always begun to philosophize." And the work certainly does evoke surprise,
though not perhaps of the kind the author intended. It has manifestly not
exercised any influence whatever upon the development of neurology, and
would not have been worth while referring to had not the author's above-
mentioned historical work acquired such widespread fame.
KddTs history of biological theories
This fame is based partly on Radl's undeniable merits as a historian: a wide
knowledge of literature, a lively style, and shrewd, often striking discern-
ment (his account of the development of Darwinism in Germany in Part II
is particularly animated and instructive); partly on his opposition to the
original Darwinism, an opposition which came into force at a date when
the old doctrine was certainly undermined, but nevertheless still officially
accepted; and partly again on the numerous philosophical digressions, some-
times witty, but more often merely odd, which are found scattered through-
out his history and which proved attractive to a generation that had wearied
of the old phylogenetical speculations without having on that account ac-
quired any other speculative foundation on which to build. Of the various
parts of the book, the first edition of Part I is the one that contains most
of the old biological ideas; in the foreword of Part II the author regrets
the far too confident belief that he had earlier entertained in an objective
science; and in the second edition of Part II, which was published last, Radl
declares that he intends to promulgate a "realistic cosmic view, such as
finds its deepest expression in Dostoievsky's novels." The work expresses
throughout, each part more extravagantly than the last, a purely panegyri-
cal enthusiasm for Aristotle, who is declared to be the unattainable ideal
as a natural philosopher, but at the same time a warm admiration for his
opponent and very antithesis, Paracelsus; and, further, the book extols
Stahl's vitalism and romantic and idealistic speculation in general, while it
disparages exact research, particularly cytology (whose methods the author
nevertheless himself employed, though not very skilfully), Darwinism, and
exact heredity-research, all of which is described as materialism. The
6li THE HISTORY OF BIOLOGY
author's subjectivism culminates in the above-quoted saying that there is no
such thing as objective science; all interest is centred upon the personalities
figuring in the history of science. This may to a certain extent be justified,
when it is a question of giving expression to the purely ideal strivings of
humanity, but when it is a question of nature, our knowledge rests undeni-
ably upon certain facts that reveal themselves equally to all. He who refuses
to admit this had best turn his back upon exact natural science for ever.
Radl has in fact done so; he is now professor of natural philosophy at Prague.
Uexkull on the life-process
On the whole, little is to be gained from the biological point of view by
becoming too deeply engrossed in the works of the modern vitalists. Some
of them have gone in for speculations about the theory of knowledge, as,
for instance, Jacob von Uexkull, who holds that only a part of the life-
process is mechanically comprehensible, while that part of it which gives
to the mechanical phenomena their " Zielsfrebigkeit" is super-mechanical and
must be referred to impulses produced by an organized natural force; me-
chanical biology is concerned with the fitting-in of every being into certain
given conditions, which give to the organism its limitations; in the world
of dew-worms tjiere exist only dew-worm conditions, while man can ob-
serve only human things. To analyse the different conditions of life and to
work out the laws governing this reciprocal action between individual and
environment is, to his way of thinking, the aim of biology. Other vitalists
have reverted to downright mysticism, and, finally, the neo-Lamarckian
school previously referred to, which is represented by Pauly and his pupils,
has tried to see in the phenomena of life, especially in evolution, expressions
for consciously operating psychical forces in the living substance. What is
common to all these different aims is the attempt to discover the difference
between animate and inanimate matter and what it is that produces the pe-
culiar character of the phenomena of life. For this purpose the methods of
physics and chemistry have proved ineffective, with the result that other
means have been sought to attain the end in view. We have already seen
that these means have proved fruitless. In such circumstances obviously the
wisest course would be: neither mechanism nor vitalism, but resignation in
face of the inexplicable. But science has not yet struck into that path, nor
is it likely to do so in the future. And fortunately too, we may well say, for
had not humanity possessed a belief in the possibility of solving the insoluble
riddles of life, there would never have been any science at all. Every delusion
that has involved an honest striving after truth has at any rate contributed
something to human knowledge, even if it is only negative, and the specu-
lations that have just been described are in this respect by no means valueless.
w
u
w
<
W
u
M
(A
w
H
2
s^ssBi«e;-^
2
CO
o
1—5
O
Q
D
MODERN BIOLOGY 613
X. The Idea of Species and Some Problems in Connexion Therewith
In a popular account of the results of heredity research published some years
ago occurs the sentence: "For the very reason of the great number of fresh
facts that modern heredity-research has brought to light, chaos prevails at
present in regard to the views on the formation of species." These words
characterize both the situation as regards the problem of species at the pres-
ent day and the causes that have brought it about. Modern heredity-research
has completely dislocated the circles drawn by the old morphological classi-
fication. Linn^eus's idea of species, it will be remembered, was essentially
genetical- he counted as many species as had been created in the begnining,
or in later years, at any rate some species created in the dawn of time in
respect of each genus, out of which the other species have since been evolved.
This idea of species could very easily be reconciled with the idealistic idea
of species which has existed since the days of Greek philosophy and which
the biology of the romantic period preferred; in those days greater attention
was paid to the idea expressed in the species form than to the question of
origin Darwinism brought the genetical idea of species once more into re-
pute To discover the origin of the different forms of life by a close compari-
son of their external and internal structure was, according to Gegenbaur,
Haeckel, and their disciples, the end of biology- thus, a natural classification
system was to be created with species based upon true relationship — species
which it is true, must be assumed to vary and overlap, but which in their
typical forms could be determined and characterized. Nevertheless, this ge-
netical idea of species rested upon an indispensable proviso — namely, that
from resemblance one could positively conclude affinity: the greater and the
more universal the resemblance, the closer the affinity. Every species, and
even every variety, had a common origin, as proved by the mutual resem-
blance between its individuals. It is this foundation for the idea of species
that modern heredity-research has undermined; it has clearly demonstrated
that very close morphological resemblance can in certain cases be due to
entirely different causes. It is not outward resemblance, but the concurrence
of hereditary factors that proves true affinity; that is to say, it is not phe-
notypical, but genotypical resemblance that determines the affinity. But
nowadays in all systematical works the species are described entirely ac-
cording to phenotypes; genotypical agreement can be ascertained only by
experimental means. And in practice, of course, this can take place only on
a small scale; the plant-geographer who makes records of localities and dis-
tribution charts at home or abroad would be stranded if, every time he sees
a form he were compelled to carry out hybridizing experiments with it in
order to establish its identity, and this applies also to the morphologist and
the systematist. In these circumstances it seems to be absolutely necessary
6l4 THE HISTORY OF BIOLOGY
to decide exactly what the categories of the system are intended to denote.
Unfortunately it has not been possible to obtain unanimity on this point;
opinions have clashed and their advocates have been very little disposed
to modify their views. We shall here cite one or two examples of these
divergences of opinion.
Lotsj and the species question
The Dutch student of the heredity problem J. P. Lotsy has taken up an
uncompromisingly genetical attitude towards the idea of species. "What is
a species?" he asks in one of his treatises. The answer is: A species in the
Linnasan sense is no species, for it comprises a large number of different life-
forms whose outward appearance bears a certain resemblance, but as to
whose origin we can determine nothing. The Linnxan species are therefore
nothing but products of the imagination, as also the Linn^an genera and
other higher classified groups; consequently they should no longer be called
species but linn^onts. Nor are the minor species into which some Linnxan
species can be divided and which remain constant, true species; these forms,
which were specially studied by the French botanist Jordan in the beginning
of the nineteenth century, should be named after him jordanonts, but they
are not species, for their internal resemblance cannot be ascertained. On the
other hand, a species is a summary of all the homozygous individuals having
the same hereditary character. "Consequently, not even all the pure-lines in
the Johanssenian sense are true species; they are so only if they are at the
same time homozygotes. And in regard to organisms with sexless repro-
duction, V, e can never know whether they are species, for that can be dis-
covered only through the analysis of hybridizing experiments." In these
circumstances it becomes a matter for doubt whether any species exist at
all in nature.
This conclusion of Lotsy's clearly proves that an insistence upon the
genetical idea of species can only lead to sheer paradox; he admits himself
that only linn^eonts and jordanonts can possess any practical systematical
significance. The whole of his reasoning is really a striking proof of the
power of language over thought; he wants what is called species to be a
genetical entity, and so he comes to a point where he does not know whether
species in his sense of the word exist at all in nature. Other students of
heredity have also taken warning from this result: thus, Heribert-Nilsson
declares that the term "species" might well be used in the form in which
Linnxus employed it; thereby, it is true, the idea of species becomes purely
morphological — "The species of classification is a phylogenetical con-
glomerate," he says — but for the genetical entities we have of course
the new nomenclature "genotype" and "pure-line." The necessity of
thus dispensing with the genetical idea of species has, indeed, been
realized by many others; Ernst Lehmann, for instance, maintained in
MODERN BIOLOGY 615
his controversy with Lotsy that the species are abstract and arbitrary
ideas and he has quoted the statements of other observers in support of
this view.
Mutations and hereditary factors
That the systematic species in the old sense of the term, comprising indi-
viduals having a certain mutual resemblance, must possess immense, even
inevitable, significance from a purely practical point of view, will at once
be realized; moreover, it has both morphological and physiological impor-
tance, in that resemblance of various kinds, both outward and inward, char-
acterize the same type of organism, from the simplest characters referred to
in the text-books to the precipitin reaction. But the species-characters —
mutual resemblance — say, as is established, nothing about mutual affinity,
and heredity research is the one branch of biology that cannot utilize the
idea of species, except under experimental control. The old speculations upon
species-phylogenesis, however, have not been given up by the representa-
tives of modern heredity-research; on the contrary, there is current among
them a veritable maze of ideas on the problem of origin. The problem has
been further complicated by a divergence of views regarding the relation
between mutations and Mendelian cleavage; while Heribert-Nilsson, for in-
stance, has found that mutations, where they exist, only cause loss of heredi-
tary factors, and on that account holds that the process of evolution can be
conceived merely as a series of reductions in the original material of genes,
the specialists in Drosophila have proved that there have existed mutations
which have given rise to positively new rudiments. The conception of the
development of life on the earth must of course be influenced by this prob-
lem. Heribert-Nilsson has resolutely followed up the consequences of his
theory that only detrimental mutations exist and has declared that a theory
of evolution is on the whole unthinkable, while other geneticists have
deemed it unnecessary to take refuge in this desperate expedient. Baur in
particular has endeavoured to create a theory of evolution by combining
the results of research work on mutations and hybridization with Darwin's
theory of selection; a similar view has been held by Morgan and his school.
Other students of heredity, again, have sought to establish a mutational
Lamarckism by assuming mutations induced by the influence of external
conditions upon the germinal plasm. Earlier attempts to prove the possi-
bility of transmitting to the offspring characters that have been experimen-
tally imparted to the parents either have, as has been mentioned before,
proved unsuccessful, or else could be given a different interpretation, as, for
instance, Kammerer's and Tower's results referred to above. Recently, how-
ever, some new observations of this kind have been carried out, certain ex-
perimentally acquired characters in animals under investigation having been
transmitted to the offspring along Mendelian lines, to this class belong, for
6l6 THE HISTORY OF BIOLOGY
example, the attempts of Little and Bagg" by passing Rontgen rays through
female rats to induce defects in the eyes and other parts of the body of their
offspring, and the experiments of Harrison^ with melanism in butterflies
induced by the introduction of metallic salts in the food. However, there
are no doubt obstinate anti-Lamarckists who have explained or will eventu-
ally explain even these results in a different way. In this connexion may
also be mentioned the recently published attempt of Professor Muller, of
Austin, Texas, to produce mutations in Drosophila by means of extreme
temperatures and Rontgen rays.
The whole of this problem of evolution is of course highly involved
and its discussion must, as far as our own times are concerned, terminate
in a number of unanswered questions. First of all, selection; that it does
not operate in the form imagined by Darwin must certainly be taken as
proved, but does it exist at all? It is obvious that by the influence of external
conditions, especially such as interfere with sudden violence, a thinning-out
of the species is possible. If, for instance, a quantity of seed from a southern
climate is sown in a northern country, the delicate plants will die, whereas
the hardy ones will live, but this selection is only a matter of relation to
cold and proves nothing as to the quality of the individuals in other respects.
But the competition between the individuals, in which Haeckel thought he
saw true selection — does it exist at all, or is it only imaginary, as O. Hert-
wig affirmed? And outside influence — has it no effect whatsoever upon the
germinal plasm and offspring of the individuals, or is there really any such
influence in the form of some kind of mutational Lamarckism?
These and many other questions it is for the future to answer. We have
now followed the history of biology up to our own times; our task is
fulfilled.
"^ Little and Bagg, Anat. Record, Vol. XXIV, 1913.
3 Proceedings of the Royal Society of London, ser. B, Vol. XCIX, 1516.
SOURCES AND LITERATURE
SOURCES
PART I
Chapters I-X
^lianus, Works, translated by Jakobs. Stuttgart, 1839.
Aristotle, Historia animalium, ed. Aubert et Wimmer. Leipzig, 1868.
— , De partibus animalium, ed. Frantzius. Leipzig, 1853.
— , De generatione animalium, ed. Aubert et Wimmer. Leipzig, i860.
— , Physica, ed. Prantl. Leipzig, 1854.
Diel, Die Fragmente der Vorsokratiker. Berlin, 1903.
Galeni opera, ed. Kiihn. Leipzig, i8ii.
— , CEuvres choisis, translated by Daremberg. Paris, 1854.
Lucretii de rerum natura, ed. Munro. Cambridge, 1893.
Plato, Timaus, selected dialogues, translated by Dalsjo. Stockholm, 1870.
(Swedish)
Plinii naturalis historia, libri VIII-XI, ed. Junius. Leipzig, 1856.
Chapters XI-XIV
Aldrovandi, Ulysses, Ortiithologia, hoc est de avibus historia, libri XII. Bologna,
1599.
Bacon, Francis, Works. London, 1847.
Belon, Pierre, La Nature et diversite des Poissons. Paris, 1555.
— , VHistoire des oyseaux. Paris, 1555.
Bruno, Giordano, Von der Ursache, dem Prin^ip und dem Einen, translated by
Lasson. Berlin, 1873.
C^esalpini, Andr^ee, Quastionum peripateticarum, libri V, ed. Vignon, 1588.
Columbi, Realdi, De re anatomica, libri XV. Venice, 1559.
Fabricii ab Aquapendente, Opera omnia anatomica et physiologica. Leipzig, 1867.
Falloppii, Gabrielis, Observationes anatomica. Cologne, 1561.
Galilei, Galileo, Opere, ed. nazjonale. Florence, 1890 ff.
Gesner, Conrad, Historia animalium. Zurich, 1551 fF.
Harvei, Guilielmi, Exercitatio de motu cordis et sanguinis in animalibus . Frank-
furt, 15x8.
— , Exercitationes de generatione animalium. London, 1551.
617
6l8 THE HISTORY OF BIOLOGY
Rondeletii, Guilielmi, De fiscibus marinis, libri XC. Lyons, 1554 fF,
Servetus, Michael, Christianismi restitutio. 1553 (ed. 1790).
Severinus, Marcus Aurelius, Zootomia Democritea. Nuremberg, 1645.
Vesalii, Andrea;, De humani corporis jabrica, libri VII. Basel, 1543.
— , Ibid., Epitome. Basel, 1543.
PART II
Chapters I-III
Descartes, Rene, CEuires pbilosophiques, ed. Aime Martin. Paris, i88i.
Helmont, J. B. van, Ortus medicina. Amsterdam, 1648.
Paracelsus, Theophrastus, Bucber und Schriften, ed. Huserus. Basel, 1589 fF.
Spinoza, Baruch, Sdmtliche IVerke, translated by Auerbach. Stuttgart, 1841.
Chapter IV
Aselli, Gaspare, De lactibus sive lacteis venis. Basel, i6i8.
Bartholin, Thomas, De lacteis thoracicis. Copenhagen, 1652..
— , Opuscula nova de lacteis. Copenhagen, 1670.
Borelli, Alfonso, De jnotu animalimn, ed. 2.. Leyden, 1685.
Glisson, Francis, Anatomia hepatis. Amsterdam, 1659.
Graaf, Reinier de. Opera omnia, ed. x. Amsterdam, 1705.
Grew, Nehemiah, The Anatomy oj Vegetables. London, 1672..
Leeuwenhoek, Antony, Opera ofnnia. Leyden, 17x1.
Malpighi, Marcello, Opera omnia. Leyden, 1687.
— , Opera posthuma. Amsterdam, 1698.
Pecquet, Jean, Experimenta nova anatomica. Amsterdam, 1661.
Perrault, Claude, Essais de physique. Paris, 1680.
Rudbeck, Olof, De circulatione sanguinis. Upsala, 1652..
— , Nova exercitatio anatotnica. Upsala, 1653.
Ruysch, Frederik, Opera omnia. Amsterdam, 1711.
Steno, Nicolaus, De musculis et glandulis. Copenhagen, 1664.
— , Elementorum myologia specimen. Florence, 1667.
— , De solido intra solidum naturaliter contento dissertationis prodromus. Florence,
1669.
Swammerdam, Jan, Bibel der Natur. Leipzig, 1751.
Vieussens, Raymond, Neurographia universalis. Lyons, 1685.
Wharton, Thomas, Adenographia. London, 1656.
Willis, Thomas, Cerebri anatome. London, 1664.
— , De anima brutorum. London, 1671.
Chapter V
Boerhaave, Hermann, Institutiones medica, ed. 3. Leyden, 1710.
Hoffman, Friedrich, Fundamenta medicina. Halle, 1703.
SOURCES 619
Hoffman, Friedrich, Medicina rationalis. Halle, 1739.
Stahl, G. E., Theoria medica vera. Halle, 1737.
Swedenborg, E., CEconomia regni animalis. London, 1740.
Sydenham, Thomas, Works, ed. Grcenhill. London, 1849.
— , Regnum animale. London, 1745.
Chapters VI, VII
Artedi, Peter, Icbthyologia. Leyden, 1738.
Bauhin, Caspar, Prodromus theatri botanici. Frankfurt, 1610.
— , V max theatri hot anici. Frankfurt, 1613.
Brunfels, Otto, Herbarum viva eicones. Strassburg, 1530 ff.
Camerarius, R.J. , Das Geschlecht der Pflanzen, translated by Mobius. Ostwald's
Klassiker der exakten Wissenschaften. Leipzig, 1899.
Cesalpino, Andrea, De Flantis. Rome, 1603.
Fabricius, J. Ch., Species insectorum. Hamburg, 1781.
Fuchs, Leonhard, Historia stirpium. Basel, 1542..
Jung, Joach., Isagoge phytoscopica. Hamburg, 1678.
Linnasus, Carl von, Amcenitates academica. Vol. I-X. Leyden, 1749 ^-
— , Classes plantarum. Leyden, 1738.
— , Fauna suecica, ed. 2.. Stockholm, 1671.
— , Fundament a botanica. Leyden, 1736.
— , Genera plantarum, ed. i. Leyden, 1742.
— , Methodus plantarum. Leyden, 1737.
— , Papers published by the Swedish Academy of Science. Upsala, 1907.
— , Sy sterna nature, ed. i, 2, 6, 10, ii. Leyden, 1735; Stockholm, 1740, 1748,
1758, 1766.
— , Juvenile Works, published by Ahrling. Stockholm, 1888.
Ray, John, Methodus plantarum, ed. i, x. London, i68x, 1733.
— , Historia plantarum. London, 1686 ff.
— , Synopsis animalium quadrupedum. London, 1693.
Tournefort, J. P. de, Institutiones rei herbaria, ed. i. Paris, 1700.
Chapters VIII-XI
Albinus, Icones ossium foetus humani. Leyden, 1737.
— , Tabula sceleti et musculorum corporis humani. Leyden, 1747.
Bonnet, Charles, Contemplation de la nature. Amsterdam, 1769.
— , Insectologie, ed. x. Amsterdam, 1780.
— , La Palingenesie philosophique. Lyons, 1770.
Buffon, G. L. L. de, Histoire naturelle. Paris, 1749 ^-
— , CEuvres completes. Paris, 1778 ff.
Camper, Petrus. Kleinere Schriften, Leipzig, 1788.
De Geer, Charles, Memoires pourservirh l' histoire des insectes. Stockholm, 1751 ff.
6xo THE HISTORY OF BIOLOGY
Hales, Stephen, Statical Essays. London, 1738.
Haller, Albrecht von, Anfangsgrunde der Physiologic des Menschen, translated
by J. S. Haller. Berlin, 1759 fF.
— , Bibliotheca anatomica. Berne, 1774.
— , M.emo!res sur la nature. Lausanne, 1746 fF.
— , On Sensible and Irritable Parts of the Body. Swedish Academy of Science,
Transactions, Vol. XLV. Stockholm, 1753.
— , Prima linea physiologia. Gottingen, 1747.
Hunter, John, Works. London, 1835.
Koelreuter, J. G., Vorlaufige Nacbricht von einigen das Geschlecht der Pfanzfn
betreffenden Versuchen. Ostwald's Klassiker, 1893.
La Mettrie, J. O. de, CEuvres philosophiques. Berlin, 1774.
Lieberkiihn, J. N., Dissertatio . . . de fabrica . . . villorum intestini. Leyden,
1745-
Lyonet, Pierre, Traite anatomique de la chenille, etc. The Hague, 1746.
Pallas, P. S., Elenchus xpophytorum. The Hague, 1766.
— , M-iscellanea %pologica. The Hague, 1766.
— , l>iova species quadrupedum, etc. Erlangen, 1778.
Reaumur, A. F. de, Memoires pour servir a V histoire des insectes. Paris, 1734 fF.
Rosel von Rosenhof, Monatliche Insektenbelustigungen. Nuremberg, 1753 fF.
Spallanzani, L., Experiences pour servir h V histoire de la generation. Geneva, 1785.
— , Programme . . . d'une ouvrage sur les reproductions animales. Geneva, 1768.
Sprengel, Conrad, Das entdeckte Geheimnis der Natur. Ostwald's Klassiker.
Leipzig, 1894.
Trembley, A., Memoires pour servir h V histoire d'une polype d'eau douce. Leyden,
1744-
WolfF, C. F., Theoria generationis, ed. z. Halle, 1774.
Chapters XII-XIV
Agardh, C. A., Text-book on Botany. Malmo, 18x9 fF. (Swedish)
Carus, C. G., Grmid^uge der vergleichenden Anatomic. Dresden, i8i8.
— , Natur und Idee. Vienna, 1861.
Darwin, Erasmus, Zoonomie, translated by Brandis. Hanover, 1795.
Goethe, J. W., Sdmtliche Werke. Cotta, Stuttgart, 1851 fF.
Herder, J. G., Sdmtliche Werke. Berlin, 1887 fF.
Hwasser, L, Essays. Upsala, 1839. (Swedish)
— , Selected Essays. Stockholm, 1868 fF. (Swedish)
Ingenhousz, J., Experiments upon Vegetables. London, 1779.
Kant, J., Gesamyneltc Schriften, Berlin, 1901.
Lavoisier, A. L., CEuvres. Paris, 1863 fF.
Nees von Esenbeck, C. G., Handbuch der Botanik. Nuremberg, i8io.
Oken, Lorenz, Naturgeschichte fur alle Stdnde. Stuttgart, 1841 fF.
SOURCES 62.1
Oken, Lorenz, Naturphilosophie. Jena, 1809.
Priestley, J., Experiments and Observations on Different Kinds of Air. London,
^775-
Saussure, N. Th. de, Kecherches chimiques sur la vegetation. Paris, 1804.
Schelling, F. W. J., Samtliche Werke. Stuttgart, 1856 fF.
PART III
Chapters I-III
Barthez, P. J., Nouveaux Elements de la science de Vhomme. Montpellier, 1778.
Bichat, F. M. X., Anatomie generale, ed. i. Paris, 1818.
— , Traite d' anatomie descriptive. Paris, 1801.
— , Traite des membranes^ ed. 2.. Paris, i8oi.
Blumenbach, J. F., Collectionis suce craniorum, etc. , decades VI. Gottingen, 1790.
— , Handbuch der Naturgeschichte. Gottingen, 1803.
— , ijber die natiirlichen V erschiedenheiten itn Menschengeschlechte , translated by
Gruber. Leipzig, 1798.
— , tjber den Bildungstrieb. Gottingen, 1791.
Bordeu, H. de, Kecherches anatomiques sur la position des glandes, ed. 1. Paris,
1799.
Cuvier, G., Le Kegne animal. Paris, 1817 fF.
— , Eecons d' anatomie cotnparee. Paris, 1799 ff.
— , Kecherches sur les as semens fossiles de quadrupedes. Paris, 18 12. fF.
Gall, F. J., Lehre uber die Verrichtungen des Gehirns. Dresden, 1805.
— , et Spurzheim, Kecherches sur le systeme nerveux, etc. Paris, 1809.
— , Sur I'origine des qualites morales. Paris, 182.2. fF.
Humboldt, A. von, Ansichten der Natur. Stuttgart, 1849.
— , Kosmos. Stuttgart, 1845 fF.
— , Versuche uber die gereixte Muskel- und Nervenfaser. Berlin, 1797.
Lamarck, J. B. de, Histoire naturelle des animaux sans vertebres. Paris, 1815 fF.
— , Memoires de physique et d' histoire naturelle. Paris, 1797.
— , Philosophie zpologique, ed. z. Paris, 1873.
— , Kecherches sur V organisation des corps vivants. Paris, i8oi.
Reil, J. C, Archiv fur Physiologie, 1796, Bd. I.
Sommerring, S. T., De corporis humani fabrica. Frankfurt, 1794 fF.
— , Uber das Organ der Seek. Konigsberg, 1796.
Vicq d'Azyr, Felix, CEuvres. Paris, 1805 fF.
— , Traite d' anatomie et de physiologie. Paris, 1786.
Chapters IV-VI
Baer, K. E. von, Uber Entwicklungsgeschichte der Tiere, Konigsberg, 182.8 fF.
— , De ovi mammalium et hominis genesi. Leipzig, 1817.
62.x THE HISTORY OF BIOLOGY
Bell, Ch., Die menschlkhe Hand, trans, by HaufF. Stuttgart, 1836.
— , Idea of a New Anatomy of the Brain, ed. Ebstein. Klassiker de Medizin, pub-
lished by SudhofF. Leipzig, 191 1.
Bernard, Claude, La Science experimental. Paris, 1878.
— , Legons de physiologie experimental . Paris, 1855.
— , Lecons sur les phenomenes de la vie. Paris, 1878.
Berzelius, J. J., Lectures on Animal Chemistry. Stockholm, 1806 ff. (Swedish,
trans, into German by Wohler)
Blainville, H. M. D. de, Traite des animaux. Paris, 1812. fF.
Fourcroy, A. F., Philosophia chemica, trans, by Sparrmann. Stockholm, 1795.
(Swedish)
Magendie, F., Legons sur les phenomenes physiques de la vie. Paris, 1836.
— , Precis elementaire de physiologie, ed. 5. Brussels, 1838.
Meckel, J. F., System der vergleichenden Anatomie. Halle, i8zi fF.
Miiller, Johannes, Bildungsgeschichte der Genitalien. Diisseldorf, 1830.
— , Handbuch der Physiologie des Menschen. Coblenz, 1837-40.
— , ijber den Bau, etc., des Amphioxus. Berlin, 1844.
— , Vergleichende Anatomie der Myxinoiden. Brllin, 1835-40.
— , Zur vergleichenden Physiologie des Gesichtssinnes. Leipzig, i8i6.
Pander, H. C, Beitrdge zur Entwicklungs geschichte des Hiihnchens im Eye. Wiirz-
burg, 1817.
Purkinje, J. E., Beohachtungen und Versuche z^^ Physiologie der Sinne. Berlin,
18x5.
— , De phanomeno, etc., motus vibratorii continui. Breslau, 1835.
— , Symbols ad ovi avium historiam. Breslau, 1815.
Rathke, H., Beitrdge z^r vergleichenden Anatomie und Physiologie. Danzig, 1842..
— , Bemerkungen uber den Bau des Amphioxus, etc. Konigsberg, 1841.
— , Untersuchungen iiber den Kiemenapparat. Riga, 1831.
Rudolphi, C. A., Entozoorum synopsis. Berlin, 1819.
— , Grundriss der Physiologie. Berlin, i8xi fF.
Scheele, C. W., Letters and Annotations, published by A. E. Nordenskiold.
Stockholm, 1891. (Partly in Swedish)
Chapter VII
Henle, J., All gemeine Anatomie. Leipzig, 1841.
— 5 Symbols ad anatomiam villorum intestinalium. Berlin, 1837.
— , Essays in Archiv fur Anatomie und Physiologie.
Kolliker, A., Entivicklungs geschichte der Cephalopoden. Zurich, 1844.
— , Entwicklungs geschichte des Menschen und der hoheren Tiere. Leipzig, i86i'.
— , Handbuch der Geivebelehre des Menschen. Leipzig, i85X.
— , Essays in Zeitschrift fur wissenschaftliche Zoologie and in Archiv fur Anato-
mie und Physiologie.
SOURCES 613
Leydig, F., Lehrbuch der Histologie des Menschen und der Tiere. Frankfurt, 1857.
— , Essays in Archiv fur Anatomie und Physiologic and in Zeitschrift fur ivissen-
schaftliche Zoologie.
Mohl, H., Grundxiige der Anatomie und Physiologic der vegetabilischen Zcllc.
Braunschweig, 185 1.
• — , Vermischte Schriften. Tubingen, 1845.
— , Essays in Botanische Zeitung.
Nageli, K., Zur Entivicklungsgeschichte dcs Pollens. Zurich, i84z.
Reichert, K. B., Be if rage %ur heutigen Entivicklungsgeschichte. Berlin, 1843,
— , Betnerhingen %ur vergleichcnden Naturforschtmg, etc. Dorpat, 1845.
— , Das Entivicklungsleben im Wirbeltierreich. Berlin, 1840.
— , Essays in Archiv fur Anatotnie und Physiologic.
Remak, R., Essays in Archiv fur Anatomic, etc.; in Cotnptcs rendus de l' Academic
des Sciences; et al.
Schleiden, M., Grund^Jige der wissenschaftlichcn Botanik. Leipzig, 1842..
— , Essays in Archiv fur Anatomic und Physiologic.
Schwann, Th., Mikroskopischc Untcrsuchungen uber die XJhereinstimmung in der
Struktur und dem Wachstum der Tiere und Pflanxen. Berlin, 1839.
Virchow, R., Die Cellularpathologie. Berlin, 1858.
— , Gcsammelte Abhandlungen. Frankfurt, 1856.
— , Essays in Archiv fur pathologische Anatomic.
Chapter VIII
Brown, R., Botanische Schriften, ed. Nees von Esenbeck. Leipzig, 182.5 ^•
Candolle, A. de. Physiologic vegetalc. Paris, i83x.
— , Theorie elementaire de la botanique, ed. t. Paris, 1918.
Du Bois-Reymond, E., Reden. Leipzig, 1886.
— , Untcrsuchungen uber tierische Elektri^itdt. Berlin, 1848 ff.
Dujardin, F., Histoire naturelle des infusoires. Paris, 1841.
Ehrenberg, C. G., Die Infusionstiere als vollkommene Organisfnen. Leipzig, 1838.
Endlicher, S., Enchiridion botanicum. Leipzig, 1841.
— , Genera plantarum. Leipzig, 1836 ff.
Fries, E., Lichenographia europaa reformata. Lund, 183 1.
— , Sy sterna mycologicum. Lund and Greifswald, i8io ff.
Hedwig, J., Theoria gencrationis, etc., plantarum crypto gamicarum, ed. i. Leip-
zig, 1798.
Helmholtz, H., Vortrdge und Reden, ed. 3. Braunschweig, 1903.
■ — , Wisscnschaftliche Abhandlungen. Leipzig, i88z.
Jussieu, A. L. de, Genera plantarum. Paris, 1789.
Leuckart, R., Uber die Morphologic der Wirbellosen Tiere. Braunschweig, 1848.
— , Uber den Polymorphismus der Individuen. Giessen, 185 1.
— , Die menschlichen Parasitcn. Leipzig, 1863.
6z4 THE HISTORY OF BIOLOGY
Leuckart, R., Essays in Miiller's Archiv.
Ludwig, K. F. W., Lehrbuch der Pbysiologie des Menschen. Heildelberg, 1851.
Mayer, R., Die Mechanik der Wdrme in Gesammelte Abha^tdlungen, published by
Weyrauch. Stuttgart, 1893.
Miiller, O. F., Vermium terrestrium et piviatilium historia. Copenhagen, 1773.
— , Animalcula infusoria. Copenhagen, 1786.
Nordmann, A., Micrographische Beitrage. Berlin, 1831.
Owen, R., On the Archetype and Homologies of the Vertebrate Skeleton. London,
1848.
— , Lectures on the Comparative Anatomy. London, 1843 fF.
— ■, On the Anatomy of Vertebrates. London, 1866.
— ■, On the Nature of Limbs. London, 1849.
Pasteur, L., Etudes sur le vin. Paris, 1868.
— , Etudes sur le vinaigre. Paris, 1866.
— , Essays in Annales de chim. et de phys. and Comptes rendus de V Academic des
Sciences.
Siebold, Th. and Stannius, H., Lehrbuch der vergleichenden Anatomic. Berlin,
1848.
Siebold, tjber die Band- und Blasemvurmer. Leipzig, 1854.
— , Wahre Parthenogenesis. Leipzig, 1856.
— , Essays in Miiller's Archiv and Zeitschrift fur tvissenschaftliche Zoologie.
Stannius, Das perpherische Nervensystem der Fische. Rostock, 1849.
Chapter IX
Biichner, L., Kraft und Stoff, ed. 7. Leipzig, i86i.
Comte, A., Cours de philosophic positive, ed. 3. Paris, 1869.
Liebig, J., Chcfnische Brief e. Heidelberg, 1844.
— , Die organische Chtmie in ihrer Anwendung auf Agrikultur und Physiologic.
Braunschweig, 1840.
— , Die organische Chemie in ihrer Anwendung auf Physiologic und Pathologic.
Braunschweig, i84x.
Mill, J. S., System of Logic, ed. 5. London, 1852..
Moleschott, J., Kreislauf des Lebens. Mainz, 1851.
Vogt, K., Kohlerglaube und Wissenschaft, ed. 4. Giessen, 1855.
Wagner, R., Menschliches Gehirn als Seelcnorgan. Gottingen, 1862..
— , Vortrag mit Diskussion im Bericht iiber die 51. Versammlung deutschcr Natur-
forscher und Arzte. Gottingen, 1854.
Chapters X-XII
Agassiz, L., An Essay on Classification. London, 1859.
— , Kecherches sur les poissons fossiles. Neuchatel, 1833 fF.
— , Essays in American Journal of Science and Arts.
SOURCES 615
Darwin, Ch., The Structure of Coral Reefs. London, 1842..
— , Geological Observations on Volcanic Islands. London, 1844.
— , A Monograph on Cirripedia. London, 1851 fF.
— , The Origin of Species, ed. i, z, 5. London, 1859, i860, 1871.
— , Variations of Animals and Plants under Domestication. London, 1868.
— , The Descent of Man. London, 1871.
— , Expression of Emotions. London, iSyz.
— , Movements and Habits of Climbing Plants. London, 1875.
— , Insectivorous Plants. London, 1875.
— , The Effects of Cross- and Self-fertilization. London, 1876.
— , On Earth-u'orms. London, 1881.
— , Life and Letters, ed. Fr. Darwin. London, 1887 fF.
Gray, A., Danviniana. New York, 1887 fF.
Hooker, J. D., Flora Tasmania. London, i860.
Huxley, T. H., Evidence as to Man s Place in Nature. London, 1863.
— , Lectures on the Elements of Comparative Anatomy. London, 1864.
— , Lay Sermons, addresses and reviews. London, 1871.
— , Life and Letters, ed. L. Huxley. London, 1900.
— , Essays in Transactions of Royal Society.
Lyell, Ch., Elements of Geology. London, 1831 ff.
— , The Geological Evidences of the Antiquity of Man. London, 1863.
Malthus, T., Essay on Population, ed. 5. London, 1817.
Quatrefages, A. de, Charles Darivin et ses precurseurs frangais. Paris, 1870.
Spencer, H., Essays. London, 1868.
— , A System of Synthetic Philosophy. London, 1870 fF.
Wallace, A. R., Contributions to the Theory of Natural Selection. London, 1870.
— , Island Life. London, 1880.
— , Darwinism. London, 1889.
Wigand, A., Der Dartvinismus. Braunschweig, 1874 fF.
Chapters XIII, XIV
Fiirbringer, M., Untersuchungen zur Morphologie und Systematik der V'ogel.
Amsterdam, 1888.
Gegenbaur, C, Grundzuge der vergleichenden Anatomie, ed. i, 2.. Leipzig, 1859,
1870.
— , Untersuchungen xur vergleichenden Anatomie der Wirheltiere. Leipzig, 1864 fF.
— , Grundriss der vergleichenden Anatomie. Leipzig, 1874.
— , Vergleichende Anatomie der Wirbeltiere. Leipzig, 1898.
— , Gesammelte Abhandlungen. Leipzig, 1911.
Haeckel, E., Die Radiolarien. Berlin, 1861, 1888.
— , Die Kalkschu'dmme. Berlin, 1871.
— , Das System der Medusen. Jena, 1879.
62.6 THE HISTORY OF BIOLOGY
Haeckel, E., Generelle Morphologic. Berlin, 1866.
— , Nattirliche Scbopjungsgeschichte. Berlin, 1868.
— , Anthropogetiie. Leipzig, 1874.
— , Gesammelte Vortrdge. Bonn, 1878.
— , Weltrdtsel. Bonn, 1903.
— , Funjz.ig J^hre Stammesgeschkhte. Jena, 191 6.
— , Entwkklungsgeschkbte emer Jugend. Published by Schmidt, Leipzig, 192.1.
— , Italknjahrt. Published by Schmidt, Leipzig, 1911.
Hubrecht, A., Dk Sdugetkrontogemse. Jena, 1909.
Mach, E., Analyse der Empfi7idungen. Jena, 1906.
— , Erkenntnis und Irrtum. Leipzig, 1906.
Miiller, F., Fur Darwin. Leipzig, 1864.
Chapter XV
Altmann, R., Die Element ar or ganismen. Leipzig, 1890.
Balfour, F., Works. London, 1884 fF.
Beneden, E. van, Kechercbes sur la maturation de I'ceuf. Paris, 1883.
Boveri, Th., Zellenstudien. Jena, 1887 fF.
Braun, A., Betracbtungen tiher die Verjungung in der Natiir. Leipzig, 1851.
Buchner, E., Essays in Bericbten der Deutscben cbemiscben Gesellscbaft.
Biitschli, O., Studien uber die ersten Entivicklungsvorgdnge der Ei^elle. Frank-
furt a. M., 1876.
— , ProtoXpa. Bronn's Klassen und Ordnungen. Leipzig, 1889.
Cajal, S. Ramon, Histologic du systhne nerveux. Paris, 1909 fr.
Engelmann, Th., Essays in Pfliiger's Archiv fur Pbysiologie.
Engler, A., Entu'ickhmgsgescbichte der Pflan'^enivelt. Leipzig, 1879.
Flemming, W., Zellsubstan^, Kern und Zellteilung. Leipzig, 1881.
Fol, H., Lehrbucb der vergleicbenden mikroskopiscben Anatomie. Leipzig, 1896.
— , Essays in IS/ihnoires de la societe de pbysique et d'bistoire naturelle. Geneva.
Goette, A., Entioicklungsgescbicbte der Unke. Leipzig, 1874.
Hansen, E. C, Essays in Mitteilungen des Carlsberg-Laboratoriums. Copenhagen.
Heidenhain, M., Plasma und Zelle. Jena, 1907 ff.
Heidenhain, R., Essays in Studien des Pbysiologiscben Instituts xu Breslau.
Hensen, V., Die Planktonexpedition. Kiel, 1890 ff.
Hertwig, O., Allgemeine Biologic, ed. 4, Jena, i9iz.
— , Das Werden der Organismen. Jena, 191 6.
— , Zur Abivebr des ethiscben, etc., Darivinismus. ]ena., 191 8.
— , Essays in Morpbologiscbe Jahrbiicber, Jenaiscbe Zeitschrift, and Zeitscbrift
fur wissenscbaftlicbe Mikroskopie.
Hertwig, R,, Lebrbucb der Zoologie. Jena, 1890.
— , Essays on the Protozoa in scientific journals.
Hertwig, O. and R., Untersucbungen %ur Morpbologie der Zelle, I-VL Jena, 1884 ff.
SOURCES 62.7
His, W., Unsere Korperform. Leipzig, 1874.
Hofmeister, W., Vergleichende Untersuchungen der Keimung hoherer Kryptogamen
Leipzig, 185 1.
Holmgren, E., Text-book on Histology. Stockholm, ic)io. (Swedish)
Kleinenberg, N., Hydra. Leipzig, 1871.
— , Essays in the Zeitschrift fur wissenschaftliche Zoologie, 1886.
Koch, R., Gesammelte Werke. Leipzig, i9ix fF.
Korschelt, E., and Heider, K., Lebrbuch der vergleicbenden Entivkklungsge-
schichte der wirbellosen Tiere. Jena, i^oi.
Kowalewsky, A., Essays in the Acta of the Imperial Academy, St. Peters-
burg.
Lang, A., Lehrbuch der vergleichenden Anatomie. Jena, 1850 fF.
Lankester, E. R., Treatise on Zoology. London, 1900.
Laveran, A., Nature par asit aire des accidents de V impaludisme. Paris, 1881.
Mobius, K., Fauna der Kieler Bucbt. Leipzig, 1865.
Nageli, C, Entstebung des Begriffes der naturbistoriscben Art. Munich, 1865.
— , Mecbaniscb-pbysiologiscbe Tbeorie der Abstammungslebre . Munich, 1884.
— , Essays in Zeitscbrift fur tvissenscbaftlicbe Botanik.
Ranvier, L., Legons d' anatomie generale. Paris, 1881.
Rollett, A., Untersucbungen uber die q^uergestreiften Muskelfasern. Vienna, 1885.
Ross, R., Researcbes on Malaria. Stockholm, 1904.
Schaudinn, P., Essays in Arbeiten aus dem Kaiserlicben Gesundheitsamte and in
Sitxungsbericbte der Konigl. Akademie der Wissenschaften. Berlin.
Schimper, A. F. W., Pflanzengeograpbie auf pbysiologiscber Grundlage. Jena, 1898.
Schwendener, S., Untersucbungen tiber den Flecbtentballus. Leipzig, i860.
— , Das mecbaniscbe Prinzip im anatomiscben Bau der Nlonocotyle. Leipzig, 1874.
— , Mecbaniscbe Tbeorie der Blattstellungen. Leipzig, 1878.
Strasburger, E., Zellbildung und Zellteilung, ed. 1-3. Jena, 1875-80.
Wiedersheim, R., Vergleicbende Anatomie der Wirbeltiere, ed. 6. Jena, 1906.
Chapters XVI-XVIII
Bataillon, A., Essays in Arcbives de Zoologie experimentale et generale.
Bateson, W., Mendel's Principles of Heredity. Cambridge, 1909.
Baur, E., Einfubrung in die experimentelle Vererbimgslehre, ed. 4. Berlin, 1919.
Bayliss, W., Principles of General Pbysiology, ed. -l. London, 1918.
Bethe, A., Essays in Arcbiv fiir die gesamte Pbysiologie, Vol. LXX.
Bunge, G., Lebrbucb der pbysiologiscben und patbologiscben Cbemie, ed. z. Leip-
zig, 1889.
Driesch, H., Analytiscbe Tbeorie der organiscben Entwicklung. Leipzig, 1904.
— , Pbilosophie des Organiscben. Leipzig, 1909.
— , Essays in Morpbologiscbe Jabrbucber and Zeitscbrift fiir tvissenscbaftlicbe
Zoologie.
62.8 THE HISTORY OF BIOLOGY
Eimer, Th., Die Entstehung der Arten. Jena, 1888 ff.
_ Galton, F., Natural Inheritance. London, 1889.
— , Essays in the Journal of the Anthropological Institute. London, 1875.
Giard, A., Contro verses transformistes. Paris, 1904.
Goebel, K., Organographie der Pflanxen. Jena, 1898.
— , Einleitung in die experimentelle Morphologie der Pflanzen. Leipzig, 1908.
Herbst, C, Essays in Arcbiv ftir Entwicklungsmecbanik.
Heribert-Nilsson, N., Experimentelle Studien uber Variabilitdt, etc. in der Gat-
tung Salix. Lund, 191 8.
Hertwig, O., Z.eit- und Streitfragen der Biologie, I, IL Jena, 1894 ff.
— , Der Kampf um die Kernfragen der Entwicklungs- und Vererbungslehre. Jena,
1909.
Hober, R., Physikalische Chemie der Zelle und der Geivebe, ed. 4. Leipzig, 1914.
Johannsen, W. L., Elemente der exakten Erblichkeitslehre, ed. x. Jena, 1913.
— , Erblichkeit. Cojsenhagen, 1917. (Danish)
— , False Analogies, Swedish by Larsson. Stockholm, 1917.
— , Biology in the Nineteenth Century. Copenhagen, 1919. (Danish)
Lehmann, A., Grund%iige der Psychophysiologie. Leipzig, i9ix.
Loeb, J., tJber den chemischen Charakter des Befruchtungsvorganges. Leipzig, 1908.
— , The Mechanistic Conception of Life. Chicago, 191 2..
Lotsy, J., Qjc est-ce quune espece. Amsterdam, 1916.
Lubbock, J., On the Senses, Instincts and Intelligence of Animals. London, 1888.
Mendel, G., Versuche uber Pflan^enhybriden. Ostwald's Klassiker der Natur-
ivissenschaft. Leipzig, 1913.
Morgan, T., The Mechanism of Mendelian Heredity. New York, 191 5.
Neumeister, R. , Betrachtungen uber das Wesen der Lebenserscheinungen. Jena, 1903 .
Nilsson-Ehle, H., Essays in the Journal of the Swedish Society of Seed-Culture
and in the journal Hereditas. Lund.
Overton, E., "Uber den Mechanismus der Resorption und Sekretion." Nagel's
Handbuch der Physiologic. Braunschweig, 1907.
Pauly, Darwinismus und Lamarckismus. Munich, 1905.
Pfeffer, W., Pflanxenphysiologie, ed. t. Leipzig, 1897 fF.
Plate, L., Selektionsprin^ip und Probleme der Artbildung, ed. 4. Leipzig, 1913.
Punnett, R. C, Mendelism. London, 191 1.
Radl, E., Geschichte der biologischen Theorien der Neux.eit- Leipzig, 1905-13.
— , Neue Lehre vom xentralen Nervensystem. Leipzig, \^ix.
Rauber, A., Essays in Morphologisches Jahrbuch, 1883.
Romanes, G., Animal Intelligence, ed. 3. London, 1888.
Roux, \V., Collected Essays. Leipzig, 1895.
— , Essays in Archiv fur Entwicklungsmechanik.
Sachs, J., Geschichte der Botanik. Munich, 1875.
— , Vorlesungen uber PJlanxenphysiologie. Leipzig, iSSi.
LITERATURE 6x9
Semon, R., Das Problem der Vererbimg erworbener Eigenschaften. Leipzig, igii.
Spemann, H., Essays in Archiv fur Etitivicklungsmechanik.
Uhlenhuth, P., Essays in Deutsche medizinische W ochenschrijt .
Verworn, M., Allgetneine Pbysiologie, ed. 5. Jena, 1909.
Vries, H. de. Die Mutationstheorie. Leipzig, 1901 ff.
— -, Essays in Berichten der deutschen botanischen Gesellschaft.
Wasmann, E., Die moderne Biologie, ed. 3. Freiburg, 1906.
Weisraann, A., Aufsdtxe uber Vererbung. Jena, 1852..
— , Das Keimplasma. Jena, 1891.
— , Uber Gertninalselektion. Jena, 1896.
— , Die Selektionstbeorie. Jena, 1909.
Wundt, W., Vorlesungen uber die Menschen- und Tierseele, ed. 2.. Hamburg, 1892 ,
LITERATURE
Boucke, Goethes Weltanschauung. Stuttgart, 1907.
Burckhardt, Geschichte der Zoologie, ed. i, 2.. Leipzig, 1907, 1911.
Carus, Geschichte der Zoologie. Munich, 187^.
Eckermann, J. P., Gesprdche mit Goethe. Leipzig, 1836.
Fries, Th., Lifine. Stockholm, 1903. (Swedish)
Gomperz, Griechische Denker. Leipzig, 1896.
Haeser, Geschichte der M.edixin. Jena, 1875.
HofFding, Geschichte der neueren Philosophie. Leipzig, 1895.
— , Modern Philosophers. Copenhagen, 1904. (Danish)
Hulth, J. M., Bibliographia linneana, I. Upsala.
Kohlbrugge, J. H. T., Historisch kritische Studien uber Goethe als Naturforscher.
Zoologische Annalen, Vols. V, VL Wiirzburg, 1913-14.
Kopp, H., Geschichte der CheTnie. Braunschweig, 1843.
Lamm, M., Sivedenborg. Stockholm, 191 5. (Swedish)
Lange, Geschichte des Materialismus, ed. i, 9. Leipzig, 1914.
May, W., Haeckel. Leipzig, 1909.
Paulsen, Fr., Kant. Stuttgart, 1899.
Radl, Geschichte der biologischen Theorien, ed. i, ±. Leipzig, 1907, 1909, 1913.
Roth, M., Andreas Vesalius. Berlin, 1891.
Sachs, J., Geschichte der Botanik, Munich, 1875.
Zeller, Die Philosophie der Griechen. Tiibingen, 1879.
INDEX
Abbe, German physicist, 548
Abelard, medieval theologian and scholar, 76
Abdallatif, Arabian zoologist, 73
Abu Sina, Persian physician and philosopher,
71
Academies, early scientific societies, 142.
Acharius, Erik, Swedish botanist, 440
Adtnographia, work on glands, by Wharton, 148
Advancement of Learning, The, by Francis Bacon,
87
itlianus, Claudius, Roman orator and natural
historian, 65, 78
Agardh, Carl Adolf, Swedish scientist, 191,
2-92-. 439
Agardh, Jacob Georg, 439
Agassiz, Jean Louis Rodolphe, American
scientist, 479, 480, 491
Airs, Waters, and Places, medical treatise by
Hippocrates, 2.6
Albertus, Magnus, mediaeval scientist, 79
Albinus, Bernhard Siegfried, German anato-
mist, 158, 159, 2.66
Albnius, 309
Aldrovandi, Ulisse, zoologist of the Renais-
sance, 56, 94, 95
Allgemeine Anatomie, by Henle, 397
AUgemeine Naturgeschichte jiir alle Stdnde,
natural history by Oken, i88
Allgemeine Naturgeschichte und Theorie des Him-
mels, by Kant, 170
Allgemeine Physiologie, by Verworn, 604
Altmann, Richard, German scientist, 536,
53S, 539, 604
Amici, Italian microscopist, 389
Analogy, term first used in comparative
anatomy by Owen, 415
Anatomy, first knowledge of, 3; earliest work
on, 13; Greek development of, 2.8; Aris-
totle's contribution to, 40; Herophilus, con-
tribution to, 51; Galen's contribution to,
62.-64; comparative, by Belon, 98; progress
in the Renaissance, 98-107; stimulus to, in
17th century, 141; progress in England, 147-
150; Malpighi's contribution to, 161;
vegetable, founded, 161-164; influence of
microscope on, 158-164; Daubenton's com-
parative, X2.J; Cuvier's pioneer work in,
333; development in the i8th century, 2.58,
351; d'Azyr's contribution to, 305; com-
parative, in France, 359; influence of
Darwinism on, 518
Anaxagoras, Greek philosopher, 13
Anaximander, early Greek philosopher, 11,
i-L, 19, 2.0, 2.2., 30, 31
Anaximandros, Greek philosopher, 453
Anaximcnes, early Greek philosopher, 12.
Ancel, French biologist, 596
Animal Life, by Brehm, 52.1
Animal Life, by Muhammed el Damiri, 73
Animalculists, school of i8th century bi-
ologists, 173, 130
Animals, relation to primitive man, 4
Animals and Plants under Domestication, by
Darwin, 471
Annalen der Chemie, published by Liebig, 408
Anselm of Canterbury, medixval theologian,
76
Anthropogenie oder Entwicklungsgeschichte des
Menschen, series of lectures by Haeckel, 515
Anthropology, development of, by Buffon,
12.6; pioneer work by Blumenbach, 307;
development of, by Retzius, 415
Antitoxins, discovery of, 597
Apathy, Stefan, Hungarian scientist, 540, 610
Appert, French chef, 430
Aquinas, Thomas, mediaeval theologian, 77
Arabs, their contribution to advancement of
science, 69-73
Archimedes, Greek philosopher and physicist,
19,69
Archiv fur pathologische Anatomie und Physiologie,
founded by Virchow, 401
Archiv fiir Physiologie, scientific journal founded
by Reil, 313
Archives de hiologie, journal published by
van Beneden, 534
Argyle, Duke of, in dispute with Huxley, 491
Aristotle, his theory of fossilization, 15; in-
fluenced by Heracleitus, 19; influenced by
Democritus, 10; his debt to Plato, 36; his
life, 33-35; his cosmogony, 36, 37; con-
tribution to biology, 38; references to, zi,
11, 45, 46, 48, 50, 53, 55-58, 61, 63, III, 114,
i^")> i35> i38> 141. 147. 161, 166, 190-193,
ii IN
198, 199, lOI, 12.1, 145, 151, 177, 181, 303,
310, 331, 334, 341, 361, 365, 430, 443, 470,
603, 609, 611
Arrehenius, Svante, Swedish scientist, 595, 598
Artedi, Peter, Swedish scientist, 109, no
Asclepiads, early Greek physicians, 15
Aselli, Gasparo, Italian physician, 141-144,
147
Astronomy, development of, by the Arabs, 69;
contribution of Cartesianism, 115
Athens, first mention of in Greek scientific
history, 13
Atomic theory, origin of, 10; Lucretius' con-
ception of, 47; development by Dalton, 49;
development by Berzelius, 49
Avenarius, Richard, Swiss philosopher, 517
Averroes, Arabian philosopher, 71, 71
Avicenna, see Abu Sina
Babylon, earliest home of civilization, 5
Bacon, Francis, 87-89, 111, 176, 111, 175, 443,
481
Bacon, Roger, mediaeval scientist, 80, 8i, 84
Bacteriology, recent developments in, 546-550
Baer, Karl Ernst von, German scientist, 363-
366, 368, 373, 376, 381, 384, 483, 494, 517,
532-
Bagg, biologist, 616
Balfour, Francis Maitland, English scientist,
530. 531
Barthez, Paul Joseph, French scientist, 345,
346
Bartholin, Caspar, professor of medicine, 144
Bartholin, Thomas, professor of medicine, 143-
147, 156
Bary, Heinrich, Anton de, German biologist,
558
Bataillon, A., French scientist, 583
Bates, Henry Walter, English scientist, 485, 487
Bateson, W., English biologist, 591
Bauhin, Caspar, professor of medicine, 194,
196
Baur, Erwin, German biologist, 591, 594, 615
Beagle, cruise of the, 461
Behring, Emil, German scientist, 548, 597, 598
Beitrdge zjir Methodik der Naturforschung, by
Wigand, 483
Beitrdge xur Optik, work by Goethe, 181
Beitrdge zur Phytogenesis, essay by Schleiden, 391
Bell, Charles, Scotch anatomist, 374-375,
377. 414
Belon, Pierre, scientist of the Renaissance, 97
Benecke, 405
Bentham, Jeremy, English political economist,
446, 459
Bergman, 180
D E X
Berlin zoological museum, founded by
Rudolphi, 353
Bernard, Claude, French physiologist, 377-
380, 434, 4S4, 596
Berzelius, Jons Jakob, Swedish chemist, 193,
354. 371-374. 383. 406. 4M. 431. 596
Bethe, Albrecht, German psychologist, 540,
601
Bible, the, quoted by Linna:us, 107
Bichat, Marie Francois Xavier, French scien-
tist, 181, 118, 191, 314, 330, 345-351, 353,
354. 359. 361. 37^. 373. 376, 379. 384. 390.
397. 398. 41^. 444. 446. 603
Bijbel der Natuure, collected works of Swam-
merdam, 168
Binary nomenclature, worked out by Linnasus,
^13
Biochemistry, modern developments in, 594-
598
Biology, development of, 3-7; Greek specula-
lations in, 10-30; influence of Plato on, 33;
contributions of Aristotle to, 37-39, 43, 44;
contribution of Galen to, 64; service of
Albertus Magnus to, 80; progress in the
Renaissance, 81-107; niodern development
initiated by Harvey, 118; contribution of
Descartes to, 115; contribution of Paracelsus
to, 137; its debt to Malpighi and Grew,
164; contribution of Leeuwenhoek to, 166;
progress in 17th century, 111-173; contribu-
tion of Linnaeus to, 111; ecological, origi-
nated by Linna:us, 115; influence of Buffon
on. 111, 118; progress in i8th century,
174-163; aid given by Kant to, 17^1; ad-
vances in beginning of 19th century, 301-
351; pioneer work by Lamarck, 316-330; im-
portance of Cuvier to, 333-341; discoveries
of Purkinje, 381; importance of Darwin to,
477; service of cytology to, 541; geograph-
ical, 558-561; modern experimental methods,
574; present position of, 603
"Biology," term created by Lamarck, 310
Bismarck, 510, 513
Blainville, Henri Marie Ducrotay de, 359-361,
390, 441, 444, 445
Blumenbach, Johann Friedrich, German anato-
mist, 306-309, 333, 334, 354, 417, 415, 450,
555
Boerhaave, Hermann, Dutch scientist, 153,
168, 183-186, 104, 106, 134, 138, 140, 345,
347. 444
Bollstadt, Albert von, see Albertus Magnus
Bonnet, Charles, Swiss scientist, 143-147, 148,
■2-95. ^97. 308, 317. 32-8, 334. 340. 415
Bordeu, Theophile de, French scientist, 345,
346, 348
I N
Borelli, Giovanni Alfonso, Italian scientist,
151-160, 174, 177, 185, 186, 130, 140, 305,
Bostrom, Kristofcr Jakob, Swedish biologist,
2-78
Botany, contribution of Theophrastus to, 45;
development by Arabs, 69; systematized
by Linnxus, 2.09, 2.11; during the i8th
century, 152.; development after Linnxus,
435-440; development in 19th century,
551-558
Bouin, French biologist, 596
Bourignon, Antoinette, religious mystic, 16S
Boveri, Theodor, German scientist, 535, 536,
581
Bowman, William, English scientist, 540
Boyle, Robert, chemist, 119, 175, 176, 179,
^54. ^75
Brandis, 383
Braun, Alexander, German botanist, 551
Breau, Jean Louis Armand de Quatrefages de,
French scientist, 484
Brehm, A., German naturalist, 511
Brewster, 2.85
Broca, 307, 415
Brogniart, 336
Brown, Robert, Scottish botanist, 391, 391,
436, 561
Brown-Sequard, Charles Edward, French
biologist, 568, 596
Brunfels, Otto, botanist, 191, 192.
Bruno, Giordano, pioneer of natural science
in the Renaissance, 49, 86, 87, 12.1, 132.
Brunner, 353
Buch, Christian Leopold von, German ge-
ologist, 455
Buchner, Eduard, German chemist, 549
Biichner, Friedrich Karl Christian Ludwig,
German thinker, 451, 511
Buckle, Henry Thomas, English historian, 459
Buffon, Georges Louis Leclerc de, French
scientist, his life, 2.19, no; his philosophy,
2.11-12.3, ^^8; history of the earth, 113, 114;
biological theory, 114-117; influence of,
118, 119; references to, 130, 131, 141, 145,
146, 170, 181, 194, 196, 197, 301, 304, 307,
316, 317, 314, 318, 331, 334, 336, 337, 416,
430. 453. 454> 484
Bunge, Gustav, physiologist, 608
Biitschli, Otto, German scientist, 534-537,
54^. 543. 545> 595
Cabanis, 317, 351
Cajal, Santiago Ram6n y, Spanish scientist,
539. 540
Calvin, persecutes Servet, no
D E X 111
Camerarius, Rudolph Jacob, professor ot
medicine, 197, 198, loc, 108, 155
Campanella, philosopher of the Renaissance,
106
Camper, Petrus, Dutch scientist, 159, 160,
180, 181, 301, 305, 307, 309, 314, 333, 334
Candolle, Augustin Pyramc de, Swiss botan-
ist, 436-438, 440, 466
Canon of Medicine, by Abu Sina, 71
Carlyle, Thomas, 561
Carolinian Institute, founded by Berzelius, 371
Carus, Gustav Carl, German philosopher, 190,
543
Cavendish, 165
Cell, smallest form of living substance, 539
Cellular physiology, Verworn's theory of, 604
Cellular theory, advanced by Wolff, 150, 151
Celsius, dean of Upsala, patron of Linnxus,
104
Celsus, 99
Cesalpino, Andrea, professor of medicine in
the Renaissance, 106, 113, 117, 161, 191-
196, 199, 107
Challenger expedition, 559
Chamisso, A. von, poet and circumnavigator.
419
Charles Darwin et scs prtcurseurs frangais, by de
Breau, 484
Chemise he Briefe, by Liebig, 450
Chemistry, first practised as a science by the
Arabs, 69; contribution of Albertus Magnus
to, 80; influence of Boyle on, 119; advance
brought about by Stahl in, 180; quantitative,
founded by Lavoisier, 165; contribution of
Berzelius to, 371; colloid, 595; ferment, 596
Chevalier, French microscopist, 389
Chevreul, 434
China, medical science of, 7
Christianismi restitutio, theological work by
Servet, no
Chromosomes, named by Waldeyer, 536
Chwolson, Russian physicist, 514
Circulation of the blood, ancient ideas of, 18,
41, 64, 65, 108, 109; ideas of the Renais-
sance, 111-114; invention of the term, 114;
discovery of, 115; theory of, 115, 116.
Classes flantarum, botanical work by Linna:us,
110, 114
Classification, early developemcnt of, 190-
101; contribution of Linnaeus to, 109, 111
Climatology, established by von Humboldt,
315
Coelum theory, developed by the Hertwigs,
5^9. 530
Colloid chemistry, modern development of,
595
IV IN
Columbus, Realdo, anatomist of the Renais-
sance, 104, 112., 116
Columella, Lucius Junius Moderatus, Roman
naturalist, 53
Comparative histology, contribution of Ru-
dolphi to, 353
Comparative morphology, advances in, 369
Comte, Isidore Auguste Marie Francois Xavier,
French philosopher, 441-447, 450, 452., 459,
493
Condillac, French philosopher, 304, 32.7, 340,
Conservation of energy theory, discovery of,
407-410
Controverses transformistes, by Giard, 567
Cope, D. E., American scientist, 568
Copernicus, Nicolaus, astronomer of the Ren-
aissance, 14, 85, 86
Correns, Carl, German biologist, 591
Corti, Italian cytologist; 396
Cosmology, Ionian conceptions, 10-13;
Pythagorean conception, 14
Cossus ligtiiperda, monograph by Lyonet, 133
Cours de philosophic positive, by Comte, 443
Cours de physiologic, by Plainville, 360
Cuenot, L., French biologist, 592.
Cusanus, Nicolaus, churchman and scholar of
the Renaissance, 84, 85, 1x2., 131
Cuvier, Georges Leopold Chretien Frederic
Dagobert, French biologist, 98, 169, iz8,
2.46, 147, X97, 198, 307, 311, 32.4, 32.8-344,
35^. 354. 355' 357-360, 361, 365. 387, 415.
416, 4^1, 415, 437, 444, 451, 454, 465, 469,
483, 567, 610
Cytology, development of, 390-405; progress
of, in 19th century, 533-544
Darstellung meines Systems, paper by Schelling,
2-75
Darwin, Charles Robert, English biologist,
life of, 461-464; geological works, 465, 466;
theory of variations, 467, 468; evolution
theory, 468-471; geography of, 469; theory
of heredity, 471; pangenesis theory, 472.;
on the descent of man, 473; sexual selection
theory, 474; general philosophy, 475;
judgments on, 476; references to, 131, 157,
191, 196, 301, 304, 316, 318, 357, 416, 447,
452-> 453. 457-460, 477-497, 501, 506, 507,
511, 514-516, 519, 515, 555, 562., 563, 566,
569. 571, 572-, 584, 587, 59O' 591. 601, 615
Darwin, Erasmus, English scientist, 194-196,
313
Daru'iniana, by Gray, 491
Darwinism, development in England and
Germany, 491
D E X
Daubenton, Louis, French anatomist, 110, 117,
118, i8i, 196, 197, 301, 303, 305, 307, 308,
333> 334
Davaine, Casimir Joseph, French physician,
547
Da Vinci, Leonardo, as pioneer in anatomy,
98. 99. 453
D'Azyr, Felix Vicq, French scientist, 304-306,
308, 318, 333-336, 371, 416
Dc cerebro anatomical work by Swedenborg, 187
De dijferentiis animalium, by Edward Wotton,
93
De Graaf, Reinier, Dutch physician and scien-
tist, 171, 171, 180, 186, 130, 363
T)c humani corporis jabrica, anatomical work by
Vesalius, 101
De r organisation dcs animaux, by Blainville, 360
Dc motu animalium, biological work by Borelli,
151-153
De naturis rerum, mediicval treatise on natural
history and science, 80
De ovi ynammalium genesi, by von Baer, 363
De piscibus marinis, by Rondelet, 96
De Plantis, botanical work by Pesalpino,
191
De re anatomica, by Columbus, 104
De trinitatis erroribus, religious treatise by
Server, no
De valvula coli, medical treatise, by Lieber-
kiihn, 159
De Vries, Hugo, Dutch biologist, 557, 587-
589. 591. 59^. 595. 603
Delage, Yves, French scientist, 583
Democritus, Greek natural philosopher, 10-11,
2-9-32-, 37, 4o> 43. 45-47, 107, III, 571
Desault, patron of Bichat, 346
Descartes, Rene, French philosopher, 113-118,
141, 148, 149, 153, 174, 187, 101, 116, 140,
310, 316, 443, 445, 599
Descent oj Man, and Selection in Relation to Sex,
by Darwin, 473
Des epoques de la nature, essay by Buffon, 113,
114
Dichogamy, discovered by Sprengel, 157
Die Lebenskraft oder der rhodische Genius, philo-
sophical work by von Humboldt, 315
Die Lebenswutider, by Haeckel, 515
Die Kadiolarien, by Haeckel, 505
Die Wahlverwandtschaften, romance by Goethe,
510
Die Weltrdtsel, by Haeckel, 514
Diogenes of Apollonia, philosopher of Ionian
school, 13
Dioscorides, Greek philosopher, 191
Dohrn, Anton, 559
Dolland, English mechanician, 389
I N
Dollinger, Ignaz, German professor of
anatomy, 363
Dominant and recessive characters, discovered
by Mendel, 591
Driesch, Hans, 580, 581, 608-610
Du Bois-Reymond, Emil, German scientist,
388, 411-413. 52-1
Diihring, E., German lecturer in philosophy,
410
Dujardin, Felix, French scientist, 42.8, 419,
508
Dumerie, pupil of Cuvier, 333
Dutrochet, 377
Dzierzon, Johann, apiculturist, 410
Ecological biology, originated by Linna:us,
2-15
Egypt, development of medical knowledge in,
6; development of biological knowledge in,
Ehrenberg, Christian Gottfried, German
scientist, 413, 42-7-42-9. 43 1
Ehrlich, Paul, German scientist, 540, 597
Eimer, Theodor, German biologist, 570, 573
Eleatic school of philosophy, 15
Embryology, advanced by Fabrizio, 105;
pioneer work by Wolff in, 2.48; progress in
19th century, 361-369; influence of Dar-
winism on, 519-531; experimental, created
by Roux, 579
Empedocles, early Greek philosopher, 16, 17,
30, 40, 44, 47, 141, 453; cosmogony of, 17-
18; scientific influence of, 18
Enchiridion, by Endlicher, 439
Endlicher, Stephen Ladislaus, Austrian botan-
ist, 438, 439
Endocrine glands, discovery of, 597
Engelmann, Theodor Wilhelm, German scien-
tist, 541
Engler, Adolf, German scientist, 561
Entdeckte Geheimnis der Natur im Bau mid in der
Befruchtung der Bliimen, Das, botanical work
by Sprengel, 156
Entelechy theory of life expounded by
Driesch, 609
Entomostraca Dania, by O. F. Miiller, 417
Entoxporum historia naturalis, work on parasites
by Rudolphi, 353
Entwickelungsgeschichte der Unke, by Goethe, 531
Entivicklungsgeschichte der Cepfjalopoden, mono-
graph by KoHiker, 401
Entwicklungsgeschichte des Menschen und der ho-
heren Tiere, text-book by Kolliker, 400
Epicurus, Greek philosopher, 46, 47, 61, 116
Epigenesis theory, introduced by Harvey, 118,
2.30; developed by Wolff, 151, 361
D E X V
Epitome, compendium of Vesalius's anatomical
works, ID!
Erasistratus of Cheos, founder of medical
school, 51
Erster Entwurf einer allgemeinen Einleitung in die
vergleichende Anatomie, by Goethe, 181
Esenbeck, Christian Gottfried Daniel Nees
von, German botanist, 189, 190, 191, 383,
436
Essais de la physique, biological essays by
Perrault, 154
Eugenics, created by Galton, 587
Eustacchi, Bartolommeo, professor of medi-
cine, 106, 158
Evolution, Anaximander's theory of, 11;
Aristotle's view of, 37, 43; as maintained by
Darwin, 468-476; expounded by Spencer,
494-497; championed by Haeckel, 506; as
defined by Giard, 567
Exercitationes de generationt animalium, by
Harvey, 117
Experimental heredity-research, 583-594
Experimental morphology, 574-583
Experimental plant-biology created by Sachs,
575
Experimental research, development of, 370-
388, 406
Expression of the Emotions in Man and Animals,
by Darwin, 475
Fabricius, John Christian, pupil of Linna;us,
117, 362.
Fabrizio, Girolamo, professor of medicine,
105, 106, 116, 117, 161
Fallopio, Gabriele, professor of medicine, 104,
105, 171
Farbenlehre, by Goethe, 181
Fauna der Kie/er Bucht, Die, by Mobius, 559
Fechner, Theodor, 599
Federley, H., Danish biologist, 594
Ferment chemistry, modern development of,
596
Fertilization, elucidation of, 541
Feuerbach, Ludwig, German philosopher, 447
Fichte, Johann Gottlieb, German philosopher,
2-73. 2.74. 2-77
Filament theory of protoplasm, advanced by
Flemming, 536, 537
Fischer, Kuno, German philosopher, 500
Flemming, Walter, Austrian cytologist, 534-
537, 539, 541, 543
Flourens, Marie Jean Pierre, French physiolo-
gist, 377
Fluxions, by Newton, 119
Fol, Hermann, Swiss scientist, 534, 535, 541,
543
vi IN
Forel, Auguste, Swiss psychologist, 6oi
Forskal, Peter, Swedish scientist, ±i6
Fourcroy, Antonie Francois de, French chem-
ist, 370-371
Fracastoro, Girolamo, Italian physician, 546
Franke, founder of Halle University, 178
Frederick II, emperor and scholar, 79
Frederick II, of Prussia, 139
Freie Wissenschaft mid freit Lehre, pamphlet by
Haeckel on education, 52.2.
Fresnel, 2.85
Fries, Elias, Swedish botanist, 439, 440
Fries, Jacob Friedrich, German philosopher,
393
Froth theory of protoplasm advanced by
Biitschli, 536, 537
Fuchs, Leonard, professor of medicine, 191,
194
Fundamenta hotanka, botanical work by Lin-
na:us, iio, xii
Fiinjzjg Jahre Sta7fim€sgeschkhte, by Haeckel,
Fur Darwin, paper by F. Miiller, 517
Fiirbringer, Max, disciple of Gegenbaur, 503,
515
Galapagos Islands, visited by Darwin, 463
Galen, Greek physician and biologist, 60-65,
69, 71, 76, 83, 91, 98-104, 108-111, 113, 137,
138. 375
Galilei, Galileo, scientist of the Renaissance,
43, 89-91, 113, 116-118, 12-1, 115, 116, 119,
130, 141, 151, 151, 154, 181, 113, 175, 373,
376, 378, 385,443,444, 515
Gall, Franz Joseph, German physician and
scientist, 310-311, 355, 445, 446
Galton, Francis, English scientist, 585-587,
589, 590
Galvani, 344
Gartner, Karl Friedrich, German scientist, 468,
584
Gas, invention of name, 139
Gassendi, Pierre, philosopher, opponent of
Descartes, 116, 154
Gay-Lussac, 449
Geber, Arabian alchemist, 70
Geer, Charles de, Swedish naturalist, 131
Gegenbaur, Carl, German biologist, 359, 469,
499-505. 511-51^. 5^^' 532-, 541. 613
Genera flantarum, by Endlicher, 439
Genera flantartwi, by Jussieu, 435
Gcnerelk Morphologic der Organismen, by Haeckel ,
417, 444, 511
Geography, established as science by von
Humboldt, 315; vegetable, created by von
Humboldt, 315
D E X
Geology, development of, 453-457
Geological Evidence on the Antiquity of Man, by
Lyell, 485
Geschichte der biologischen Theorien der Neuzeit,
by Radl, 498, 610
Geschichte der Botanik, by Sachs, 551, 575
Geschichte des Materialistnus, by Lange, 450
Gesner, zoologist of the Renaissance, 56, 93,
94> 19^
Giard, Alfred, French biologist, 567
Gladstone, in controversy with Huxley, 490
Glisson, Francis, professor of medicine, 147,
148, 149, 156
Goebel, Karl Eberhard, German scientist, 577
Goethe, Johann Wolfgang von, 117, 193, 115,
138, 147, 173, 174, 176, 179-185, 187-193,
196, 198, 310, 330, 341, 356, 359, 367, 368,
38c^3S4, 387, 393, 441, 494, 501, 511, 511,
514, 515, 510, 516, 551, 551, 578
Goette, Alexander Wilhelm, German scien-
tist, 531
Golgi, Camillo, Italian scientist, 539
Granule theory of protoplasm, advanced by
Altmann, 536, 538
Grassi, Giovanni Baptista, Italian scientist,
549
Gravitation, discovered, 119
Gray, Asa, American botanist, 491, 491
Greece, first to develop natural science, 5, 8;
national character of, 8; religion of, 8, 9;
educational system of, 9; early scientists of,
10-19; natural philosophy of, 11, 13;
medical science of, 15, 16
Grew, Nehemiah, English physician and
scientist, 161-164, ^97> ^°°' ^^4' 39°
Grisebach, August Heinrich, German scien-
tist, 561
Grundriss der Physiologic, text-book by
Rudolphi, 354
Grinid'zuge der wissenschajtlichen Botanik, text-
book by Schleiden, 393
Haeckel, Ernst Heinrich, German biologist,
life of, 505-508; works of, 508-510, 511,
515,519; philosophy of, 511; morphology
of, 5 11; mechanical interpretation of nature,
513; biogcnetical principle of, 516-519;
references to, 117, 131, 185, 318, 319, 330,
358. 359. 393. 399. 403. 413. 417. 444-447,
469, 473, 491, 499, 500, 511 513-52.6, 519,
531-534. 542-, 544, 551-554. 562., 564. 566,
568, 570, 571, 573, 578, 579, 598, 599, 603,
608, 613, 616
Hcemastaticks, scientific treatise by Hales, 154
Hales, Stephen, English botanist, 119, 151-
2-54. 2-65, 370
I ND
Haller, Albrecht von, Swiss scientist, 134-138,
144, 149, 195, 303, 349, 364, 370, 374, 379
Hamm, Dutch biological student, 166
Haiidbuch der Gewebelehre des Menschen, by K61-
liker, 400
Handbuch der Physiologic des Menschen, by Miil-
ler, 384
Hansen, Emil Kristian, Danish biologist, 548
Hansen, F. C. C, Danish scientist, 541
Harrison, English biologist, 616
Harting, 405
Hartmann, Eduard von, German philosopher,
Hartmann, M., pupil of Schaudinn, 550
Harvey, William, English scientist, 61, in,
114, 118, 111, 114, 115, 141, 141, 144-147,
149, 151, 158, 161, 169, 170, 139, 151, 158,
361, 379, 430, 541, 541
Hasselqvist, F., pupil of Linnxus, 116
Hedwig, Johann, Hungarian botanist, 439
Heer, Oswald, Swiss scientist, 561
Hegel, Georg Wilhelm Friedrich, German
philosopher, 178, 198, 393, 450, 516, 551, 556
Heidenhain, Martin, German scientist, 535,
538
Heidenhain, Rudolf, German scientist, 535,
539
Hellwald, F. von, German geographical
writer, 511
Helmholtz, Hermann Ludwig Ferdinand,
German scientist, 388, 408-410, 411, 413,
5^6, 541
Helminthology, advanced by Siebold, 410
Henle, Jacob, German scientist, 388, 397-398,
400, 401, 546, 608
Hensen, Victor, German scientist, 559
Heracleitus of Ephesus, early Greek philoso-
pher, 19, II, 18, 44
Herbarum viva eicones, botanical work by Brun-
fels, 191
Herbst, Curt, German scientist, 581
Herder, Johann Gottfried, German poet and
student, 171, 173, 179-181, 331
Heredity, modern experiments in, 583-594
Heribert-Nilsson, Swedish naturalist, 588, 614,
615
Herophilus, teacher in the Museum at Alexan-
dria, 51
Hertwig, Oscar, German anatomist, 470, 503,
518, 519, 533-535, 541, 543, 567-570, 574,
579, 580, 616
Hertwig, Richard, German zoologist, 390,
52-9. 533. 534. 545. 546, 582-
Heyse, Paul, 507
Hildegard of Bingen, nun and author of
Physica, 78
EX Vll
Hindu science, 7
Hippo, early Greek naturalist and philosopher,
13
Hippocrates, Greek pioneer of medical science,
16-18, 175, 177, 361
His, Wilhelm, Swiss scientist, 405, 518, 530,
531, 540, 578
Histoire des oyseaux, by Belon, 98
Histoire des sciences de I'organisme, by Blainville,
360
Histoire nature lie, by Buff on, 110
Histoire naturelle de I'dme, work by La
Mettrie, 139
Histoire naturelle des animaux, by Buffon, 114
Histoire naturelle des animaux sans verterbres, by
Lamarck, 310
Histoire naturelle des crustacees, by Milne-
Edwards, 415
Histoire naturelle des estranges poissons marins,
monograph by Belon, 97
Histologie du systeme nerveux, by Ramon y Cajal,
540
Historia animalium, biological treatise by
Gesner, 93, 94
Historia plantarum generalis, botanical work by
Ray, 199
Historia Stirpium, botanical work by Fuchs,
191
History of Philosophy of Later Times, by Hoff-
ding, 441
Hobbes, Thomas, 116
Hoffding, 516, 599
Hoffmann, Friedrich, professor of medicine,
176-178, 185-188, 106, 140
Hofmeister, Wilhelm, German botanist, 558
Hohenheim, Theophrastus, see Paracelsus
Holbach, 168
Holmgren, Emil, Swedish scientist, 541
Homme machine, L' , polemic by La Mettrie, 139,
140
Hofno duplex, anatomical treatise by Dau-
benton, 118
Homology, term created by Owen, 415
Hooker, Joseph Dalton, English botanist,
463, 491-495, 561
Horses of Elberfeld, 601
Hubrecht, A. A. W., Dutch scientist, 504
Humboldt, Alexander von, German scientist,
314-316, 334, 344, 351, 397, 417, 436, 441,
447.453.558.560
Hume, David, Scotch philosopher, 195, 491
Hunter, John, Scotch anatomist, 160-161, 309
Hutton, James, Scotch geologist, 454, 456
Huxley, Thomas Henry, English biologist,
417, 446, 473, 481, 485, 488-495, 501, 517,
544
VIU INDEX
Huygens, 155
Hwasser, Israel, Scandinavian philosopher,
191, 193, 3ii, 373
Hjdrachna, by O. F. Miiller, 42.7
Ibn-Rushd, see Averroes
Idea of a New Anatomy of the Brain, by Bell,
375
Ideen 7ji einer Philosophie der Natur, philo-
sophical work by Schelling, ^74
Ideen 'Zjif Philosophie der Geschichte der lAensch-
heit, philosophical work by Herder, 2.73
Idioplasma theory, developed by Nageli, 556
Ilmoni, Immanuel, Finnish philosopher, 193
Infusionstierchen a Is vollkommene Organismen, by
Ehrenberg, 417
Ingemarsson, Nils, father of Linnaeus, 103
Ingenhousz, Jan, Dutch scientist, 2.66, 2.67, 370
Institutiones medica, by Boerhaave, 185
Internal secretion, process of, discovered by
Bernard, 596
Ionic philosophers, earliest of Grecian scien-
tists, 10
Isagoge phytoscopica, handbook of botanical
study by Jung, 195
Isis, German journal of science, 2.87
Israelites, their conception of nature, 6
Jacobsen, founder of the Karlsberg laboratory,
Jacobson, Ludwig, Swedish biologist, 405, 4x4
548
Jager, C, German naturalist, 511
Janssen, inventor of microscope, 158
Johanssen, Wilhelm Ludwig, Danish biologist,
89, 461, 565, 568, 57i, 588, 589, 591
Jordan, French botanist, 614
Joule, J. P., English physicist, 408-410
Jung, Joachim, botanist, 194, 195, 199, 109
Jussieu, Antoine Laurent de, French botanist,
435. 436
Jussieu, Bernard de, French botanist, 435, 436
Kalkschwdmme, Die, by Haeckel, 509
Kalm, Per, Finnish biologist, 2.16
Kammerer, 569, 615
Kampf der Teile im Organismus, Der, by Roux,
566
Kant, Emmanuel, German philosopher, 2.69-
^75. 334. 340. 393. 447. 491. 5^6, 607
Karlsberg laboratory, 548
Kepler, 52.5
Keplerbund, formation of, 5x5
Kielmayer, Karl Friedrich, German biologist,
331
Kleinenberg, Nicolaus, German biologist, 532.,
578
Klingenstierna, Samue , Swedish physicist,
389
Koch, Heinrich Hermann Robert, German
scientist, 540, 547-548. 55°
Koelreuter, Joseph Gottlieb, German plant-
physiologist, 154, 155, 156, 468, 584
Kohlbrugge, German biologist, 453
Kolliker, Rudolf Albert, Swiss biologist, 388,
400, 401, 418, 481, 49i, 499, 505, 540, 551,
570, 587
Kosmos, cosmology by von Humboldt, 316
Kowalewsky, Alexander, Russian biologist,
5^9
Kraft und Staff, popular scientific work by
Biichner, 451
Kreislauf des Lebens, by Moleschott, 450
Kristall-Seelen, by Haeckel, 515
Kritik der reinen Vernunft, philosophical work
by Kant, 170
Lacaze-Duthiers, Felix Joseph Henri, French
scientist, 415
Lamarck, Chevalier de, French scientist, 12.8,
Z46, 147, 196, 316-336, 339, 340, 343, 351,
357-361. 416, 430, 437. 438. 446. 456-458.
464, 467, 484, 489, 511, 555, 564
La Mettrie, Julien Offroy de, French scientist,
138-2.43, 145, i6o, z68, 177, 2.80, 2.95, 314,
373
Lang, Arnold, Swiss anatomist, 5x9, 592.
Lange, Albert, 413, 450
Lankester, Edwin Ray, English scientist, 530
Latour, Charles Cagniard de, French scientist,
431
Laveran, Alphonse, French physician, 549
Lavoisier, Antoine Laurent, French scientist,
180, 2.65-2.67, Z75, 313, 319, 334, 335, 370,
407. 431
Leche, W., 333
Legofis sur l' anatomie comparie, by Cuvier, 334
Lectures on Animal Chemistry, by Berzelius, 372.
Leeuwenhoek, Antony van, Dutch scientist,
158, 164-166, 170, 171, 180, 186, 2.30, 350,
389, 42.6
Lehmann, Alfred, pyschologist, 599, 601
Lehmann, Ernst, 614
Lehrbuch der Botonik, by von Esenbeck, 189
Lehrbiich der Histologic, by Leydig, 401
Lehrbuch der Naturphilosophie, by Oken, 2.88
Lehrbuch der Physiologic, by Ludwig, 413
Lehrbuch der vergleichenden Anatomic, by Siebold
and Hannius, 417, 418
Leibniz, Gottfried Wilhelm, philosopher, 12.7,
12.8, 130, I4Z, 174, 177, 183, ii5, i2.i, 1x3,
1x5, i44, Z45, 148
Leucippus, early Greek philosopher, 2.0
INDEX
IX
Leuckart, Karl Georg Friedrich Rudolf, Ger-
man zoologist, 411, 42.2., 450, 563
Leydig, Franz, German cytologist, 401, 499
Liberalism, in the 19th century, 458
Lieberkiihn, Johann Nathanael, German
scientist, 158, 159, 389
Liebig, Justus, German chemist, 431, 448, 449,
450
Linnaeus, Carl, Swedish scientist, life of, 2.03-
io6, L09; fame of, 2.06; philosophy of, zo6,
107, 115, L16; as a systematician, 2.07, 2.10-
115; pupils of, XI6-2.I7; references to, 184,
191, 193, 195, 198, 2.01, 12.0-2.2.2., 2.16, 2.^7,
130, 2.32., 135, 155, 2.78, z8l, 192., 303, 308,
315, 3^6, 32.7, 339, 340, 346, 417, 435-440,
445,446, 514, 546, 560, 575,613
Linne, Carl von, son of Linnasus, 2.05, 2.06
Linnean Society, founded, lo6
Lister, 547
Little, biologist, 616
Locke, English philosopher, 175, 317
Loeb, Jacques, member of Rockefeller Insti-
tute, 581, 583, 605, 606
Loffler, F. J. S., German scientist, 548
Lofling, P., pupil of Linnasus, ii6
Lotsy, J. P., Dutch biologist, 614
Lotze, German philosopher, 450
Lovcn, Sven Ludwig, Swedish biologist, 419,
413, 414, 543
Lubbock, John, Lord Avebury, 601
Lucretius, Roman philosopher, 47-49, 87,
12.6, 154
Ludwig, Karl Friedrich Wilhelm, German
physiologist, 413, 414, 535
Lyell, Charles, Scotch geologist, 453, 455-457,
464, 470, 485
Lymphatic system, discovery of, 144, 145
Lyonet, Pierre, French biologist, 133
Lyser, Michael, professor of medicine, 144
Mach, Ernst, Austrian scientist, 52.6, 52.7,
605
Magendie, Frangois, French professor of
medicine, 375-389, 384, 386, 406, 444
Maillet, de, 484
Maistre, Joseph de, French author and posi-
tivist, 441
Malpighi, Marcello, professor of medicine,
M9i 159-164, 166, 185, 188, 196, ioo, 2.14,
2.30, 2.51, 350, 361, 368, 373, 389, 390
Malthus, Thomas Robert, English economist,
453, 460, 464, 470, 486
Manuel d' actinologie et de :(oophytologie, by Blain-
ville, 360
Marx, Karl, German socialist, 447
Materialism, in the i8th century, 2.68, 2.69
Mayer, Julius Robert, 407-410, 448
Mechanical explanation of life-phenomena,
151-158
Mkhanique des animaux, by Perrault, 154
Mechanische Prinzip im anatomischen Bau der
Monokotylen, Das, by Schwcndener, 557
Mechanisch-physiologische Theorie der Abstam-
mungslehre, by Nageli, 555
Mechanism, modern theory of, 603
Meckel, Johann Friedrich, German scientist,
355-359. 365, 367, 393. 516, 549
Medicine, beginning of science of, 4, 5; de-
velopment in Egypt, 6, 50; development in
China, 7; development in Greece, 2.5-2.6;
contributions of Herophilus, 51, 51; de-
velopment by Arabs, 69; state of, in By-
zantine Empire, 74; progress in the Renais-
sance, 99-107; influence of Paracelsus on,
137; development in 17th century, 141-150
Meisenheimer, J., German biologist, 597
Memoires de physique et d'histoire naturelle, by
Lamarck, 318, 319, 310
M-cmoires pour servir i I' histoire des insectes, by
Reaumur, 2.31
Mendel, Gregor, Austrian biologist, i55, 156,
469, 471, 490, 557, 588, 590^592., 604
Mendel, Johann, see Gregor Mendel
Mendelism, spread of, 592.
Menscbenschopfung und Seelensubstanz., lecture by
Wagner, 450
Mesmer, 344
Methodus plantarum, botantical work by Lin-
naeus, 109
Methodus platitarum, botanical work by Ray,
199
Metschnikoff, Ilja, Russian biologist, 508, 598
Micella theory of cells, developed by Nageli,
554
Michaelis, Caroline, wife of Schelling, 2.74
Microbiology, developed in 19th century, 42.6-
435, 544-550
Micrographie, by Mohl, 391
Micropyle, discovered by Leuckart, 442.
Microscope, invention of, 158; improved by
Leeuwenhoek, 165; development in 19th
century, 389
Microscopical anatomy founded by Malpighi,
159
Microscopy, science of, founded by Bichat,
347; advances made in 19th century, 389,
405
Microtome, introduced by His, 405
Mikrographische Beitrdge, by Nordmann, 412.
Mikroscopische Untersuchutigen iiber die Uberein-
stitnmung in der Struktur und dem Wachstum
der Tiere und Pflanzen, by Schwann, 394
X I ND
Mill, James, English philosopher, 446
Mill, John Stuart, English philosopher, 447,
459
Milne-Edwards, Henri, Belgian scientist, 415,
434
Mirandola, Pico de, Italian mystic, 131
Mirbel, Charles Frangois, French botanist, 390
Mobius, Karl August, German scientist, 559
Mohl, Hugo, German scientist, 391, 396, 449
Mohr, O. L., Norwegian biologist, 594
Moleschott, Jacob, Dutch physiologist, 449,
45^
Monad theory, ix8
Monatliche Insectenhelustigungen, natural-history
work by von Rosenhof, 2.33
Monet, Jean Baptiste Pierre Antoine de, set
Lamarck
Monism, philosophical theory, 516
Monist League, formation of, 52.5
Morseus, father-in-law of Linnaeus, 104
Morgan, Thomas Hunt, American biologist,
581, 591, 593, 615
Morphology, development in the 19th century,
414; influence of Darwinism on, 518-558
Muhammed el Damiri, Arabian zoologist, 73
Muller, American biologist, 616
Miiller, Fritz, German biologist, 469, 516-518
Muller, Heinrich, 405
Muller, Johannes Peter, German scientist, lyz,
2-85, 375. 382.-388, 394-396, 398, 400, 402.,
403, 406, 409, 411-414, 417, 4ii, 42.4, 449,
499. 505. 508, 516, 572.
Muller, Otto Frcderik, Danish scientist, 4x6,
4^7
Museum fiir Volkerkunde, founded by Vir-
chow, 402.
Mutation theory, advanced by de Vries, 588
Mystical speculation in science, 1 31-140
Nageli, Karl Wilhelm, Swiss botanist, 396,
42-9> 535. 547, 551-557, 563, 565, 568, 570,
575, 591
Napoleon, 32.6, 331, 333
Nathorst, Alfred, Swedish scientist, 561
Natur und Idee, by Carus, 2.90
Natural History, by Pliny, 53, 54
Natural History of Creation, by Haeckel, 591,
607
Natural Inheritance, by Galton, 585
Natural philosophy, character of, in i8th cen-
tury, 186-198, 301
Natural selection, as adduced by Darwin, 464,
470, 471; upheld by Weismann, 566; at-
tacked by Hertwig, 568; defended by Plate,
571; as applied by Roux, 579; present-day
view of, 616
E X
Nature et diversites des poissons. La, monograph
by Belon, 97
Natiirliche Schopfungsgeschichte, series of lectures
by Haeckel, 515
Needham, English microscopist, 430
Neo-Darwinism, 561-573
Neo-Lamarckism, 561-573
Nernst, 595
Nettesheim, Heinrich von, German mystic,
132-
Neut Lehre vom Zftitrakn Nervensystem, by Radl,
610
Neumeister, R., physiologist, 608
Neurologis universalis, work on the nervous
system by Vieussens, 150
New M.a?nf?ial Species from the Kodentia, biologi-
cal work by Pallas, 163
Newton, Isaac, English scientist, 43, 91, 119-
131, 155, 177, 181, 113, 153, 170, 175, 176,
181, 1S3, 376, 389, 476
Nietzsche, 561
Nilsson-Ehle, H., Swedish biologist, 591
Nissl, Friedrich, German physician, 540
Nordmann, Alexander von, German biologist,
42-^-, 4^3
Novum Organum, by Francis Bacon, 87
Observationes anatomicce, anatomical work by
Fallopio, 104
CEconomia natura, treatise by Linnaeus, 115
Oken, Lorenz, German philosopher, 184, 187-
189, 191, 354, 366, 383, 387, 400, 415, 416,
489, 501, 511, 515, 551
Old Testament, record of Israelitic conception
of nature, 6
On Air and Fire, chemical treatise by Sheele,
164
On Coral Reefs, by Darwin, 465
On Diet, early Greek medical treatise, 19
On Nature, philosophical poem by Anaxi-
mander, 11
On the glands, by Bordeu, 345
On the Habits of Animals, by Claudius /Elianus,
On the History of Animals, biological work by
Aristotle, 37
On the Nature of Things, philosophical poem by
Lucretius, 47
On the Origin of Species by Means of Natural
Selection, by Darwin, 465, 471, 473
On the Parts of Animals, biological work by
Aristotle, 37
On the Relations of Numbers in connexion with the
Movements of Animals, treatise by Muller, 383
On the Reproduction of Animals, biological work
by Aristotle, 37
I ND
On the Soul, biological work by Aristotle, 37
On Volcanic Islands, geological treatise by
Darwin, 465
Ophthalmascope, invented by Hclmholtz, 409
"Opsonins," discovery of, 598
Opuscula anatomka, treatise by Eustacchi, 106
Ordo plantarum, botanical work by Rivinus,
195
Organic chemistry, 370, 371; development of,
406
Organische Bewegung in ihrtm Zusammenhange
mit dem Stojfwechsel, essay on energy by
Mayer, 408
Organographie, by de Candolle, 438
Origin of species, Darwin's theory of, 465
Orsted, 2.77
Osteographie, by Blainville, 360
Osteology, Greek study of, z8
Overton, Charles Ernest, English scientist,
595
Ovists, school of 18th-century biologists,
17}. 2.30
Owen, Richard, English anatomist, 414-417,
4x8, 441, 478, 479, 489, 502.
Palaeontology, contribution of Steno to, 157;
pioneer work of Cuvier, 337
Pallas, Peter Simon, German scientist, 1.60,
^62., 353
Pander, Heinrich Christian, German scientist,
363, 368-369, 517
Pangenesis theory of Darwin, 471
Paracelsus, 133-139, 177, 180, 2.69, 2.79, 611
Paramirum, work on science by Paracelsus, 135,
136
Parasitic research, pioneer work by Rudolphi
in. 353
Parmenides, Greek philosopher of Eleatic
school, 15-18, zo
Parthenogenesis, discovered by Bonnet, 144
Pasteur Institutes, establishment of, 432.
Pasteur, Louis, French scientist, 432.-435, 449,
5V. 546, 548. 549. 554> 571. 597
Paulus of /Egina, Byzantine physician and
author, 74, 75
Pauly, August, German biologist, 571, 6iz
Pecquet, Jean, medical student, 143, 144
Perigenesis der Plastiduh, Die, by Haeckel, 519
Peripatetic Problems, by Cesalpino, 113, 114
Perrault, Claude, French physician and archi-
tect, i53-i55> 158, 174. 177. MO. 305. 310
Petersen, C. G. J., Danish scientist, 559
Pfeffer, Wilhelm, German botanist, 576, 578,
589. 595
Pflaniengeographie auf physiologischer Grundlage,
by Schimpfcr, 561
EX XI
Phagocyte theory, produced by MctschnikofF,
598
Pharmacology, beginnings of, 5; development
by Arabs, 69
Phonological biology, originated by Linnaeus,
ZI5
Philosophica botanica, by Linnaeus, Z14, 2.82., 437
Philosophie chimique, by de Fourcroy, 371
Philosophic zpologique, by Lamarck, yio, 32.3,
314, 3i6, 330
Philosophy, origin of term, 11
Phlogiston, chemical theory invented by Stahl,
179, 180, x64, z65
Physica, mediaeval treatise on natural history,
78
Physiologic vigetale, by de Candolle, 438
Physiologus, medixval treatise on natural his-
tory, 78
Physiology, early ideas of, 3, 4, 5; Greek de-
velopment of, x-j, 18; Aristotle's conception
of, 41; aid of Paracelsus to, 137; contribu-
tion of Haller to, Z35, 136; evolutional, 578
Pinax theatri botanici, botanical work by
Bauhin, 194
Plate, Ludwig, German biologist, 572., 573
Plato, Greek philosopher, 19, 31-33, 41, 48,
59, 61, 65, 76, IZ3, 190, z69, 611
Pliny, Roman natural philosopher, 53-57, 65,
73, 80, 94, 191, XXi
Plotinus, founder of neo-Platonism, 66
Politia natura, treatise by Linnaeus, 2.15
Polo, Marco, 81
Positive philosophy, system developed by
Comtc, 443
Positivism, as a conception of nature, 441
Pouchet, Felix Archimede, French scientist,
433. 434. 479. 571
Precipitin reaction, discovered by Uhlenhuth,
598
Preformation theory, developed by Swammer-
dam, 170, 12.5, x3o; Bonnet's conception,
Z45, 2.48, 149-151, 3 6z
Prenant, Auguste, French scientist, 541
Priestlev, John, English chemist, i8o, 164, 165
Principle of Population, The, by Malthus, 460
Principles of Geology, by Lyell, 456
Pringsheim, Nathaniel, German botanist, 558
Printing, its service to the spread of scientific
knowledge, 91, 94, 99
Prodromus, botanical work by Bauhin, 194
Prodrotnus systematis naturalis regni vegetahilis,
by de Candolle, 437
Prometheus, poetical work by Goethe, z8o
Protagoras, founder of Sophist philosophy, 14
Protozoa, investigations of, 545
Prowazek, S., pupil of Schaudinn, 550
xii I N
Psychiatry, reformed by Reil, 311
Psychology, animal, 599, 601, 601; insect, 600
Psycho-physics, discovered by La Mettrie, 2.41
Punnett, R. C, English biologist, 591
Purkinjc, Johannes Evangelista, Czech scien-
tist, 2.85, 36}, 380-383, 417, 414, 575
Pyrrho of Elis, Greek philosopher, 50
Pythagoras, early Greek scientist, 13, 14, 16,
17, 32-> 12.3
Pythagorean philosophy, 13, 14
Quastiones naturales, philosophical work by
Seneca, 107
Rabl, Karl, German scientist, 544
Radl, Emmanuel, Hungarian biologist, 470,
477. 498. 610-611
Ranvier, Louis Antoine, French scientist, 540,
541
Rathke, Martin Heinrich, German scientist,
363, 364, 366-368, 382., 384, 386, 489
Rauber, August, German anatomist, 578
Ray, John, scientist, 198, 199-2.01, 108-111,
2-2-3. 435. 436
Reaumur, Rene Antoine Ferchault de, French
scientist, 131, 131, 143, 490
Kecherches sur les ossemens fossihs, by Cuvier,
337, 338
Recherches sur I' organisation des corps vivants,
biological work by Lamarck, 310, 311, 313,
3M
Redi, Francesco, Florentine physician, 430
Kegne animal, anatomical work by Cuvier, 333,
339
Reichert, Karl Bogislaus, German cytologist,
398-400
Keign of Law, The, by the Duke of Argyle,
491
Reil, Johann Christian, German scientist, 311-
314. 354. 355, 37^
Reincke, J., German botanist, 515, 608
Rejuvenation experiments, 597
Remak, Robert, German scientist, 388, 399-
400, 517
Retzius, Anders Adolf, Swedish biologist, 193,
307, 414, 415
Retzius, Gustaf, son of Anders Retzius, 540,
541, 543
Rivinus, Augustus Quirinus, professor of
medicine, 195, 196, 100, 108
Rollett, Alexander, German scientist, 541
Romanes, George John, English psychologist,
599. 601
Romanticism, as a conception of nature, 441
Rondelet, Guillaume, biologist of the Renais-
sance, 96, 97
D E X
Rosenberg, H. O., Swedish biologist, 594
Rosenhof, August Roesel von, German natural-
ist, 133
Ross, Ronald, English scientist, 549
Rothman, teacher of Linnaeus, 103
Rousseau, J. J., 459
Roux, Wilhelm, German biologist, 531, 566,
578-580, 604, 608, 610
Royal Society, founding of, 147
Royal Society of Medicine in France, founding
of, 301
Rudbeck, Olof, 145-147, 187, 159
Rudbeckius, Johannes, bishop, 145
Rudolphi, Carl Asmund, German anatomist,
352--355. 358. 375, 38^, 384-386, 417, 418,
411, 413, 431
Ruysch, Frederic, Dutch physician and scien-
tist, 170, 171, 185
Sachs, Julius, German botanist, 551, 575-578,
580, 587, 605
Saint-Hilaire, Etienne GeofFroy, French bi-
ologist, 196-198, 305, 317, 331, 333, 336,
341. 343, 355. 356, 358, 416, 437, 441, 483,
511, 610
Saint-Hilaire, Isidore, French biologist, 198
Sakarja ben Muhammed, "el Kasvini," natural
historian of Arabia, 73
Sars, Michael, Norwegian zoologist, 367, 419,
4^3 _ ^
Saussure, Nicholas Theodore de, Swiss scien-
tist, 166, 167, 575
Schaudinn, Fritz, German scientist, 550
Scheele, Carl Wilhelm, German scientist, 180,
164, 370
Schelling, Friedrich Wilhelm Joseph, German
philosopher, 173, 174-177, 178, 181, 183,
186, 188, 189, 191-193, 197, 313, 330. 342-,
354. 356. 393. 511. 513. 52-6, 553
Schiller, 381
Schimper, Andreas Franz Wilhelm, Swiss
scientist, 561
Schimper, Karl Friedrich, German botanist,
551
Schleiden, Matthias Jacob, German scientist,
39^-394. 395. 396,449. 505, 552-
Schmidt, J., German scientist, 559
Schultze, Max Johann Sigismund, German
biologist, 403-405, 500, 508, 533
Schwann, Theodor, German professor of
anatomy, 388, 394^395. 398, 399, 431,
433
Schwendener, Simon, Swiss biologist, 557
Science de I'hotnme, by Barthez, 345
Scotus, Michael, translator of Aristotle, 79
Sedgwick, Adam, English geologist, 461
IN
Selektioiuprinz.ip und Probhnie der Artbildung,
by Plate, 572.
Semon, Richard, German biologist, 570
Semper, Karl, German biologist and explorer,
560
Seneca, Roman philosopher, 107
Serology, modern development of, 597
Server y Reves, Miguel, 109-112.
Servetus, Michael, see Server y Reves
Severino, Marc' Aurelio, Italian anatomist,
106, 107
Sexual selection theory of Darwin's, 474
Sexuality of plants, 197, 198
Shaksjxre, 602.
Siebold, Karl Theodor Ernst von, German
biologist, 417-42.1, 42-1, 419
Smith, William, English palaeontologist, 454,
455
Socrates, Athenian philosopher, 31, 169
Solander, Daniel, pupil of Linnarus, 117
Sommerring.SamuelThomas, German scientist,
309, 310, 353
Sophist school of philosophy, 2.3, 30
Spallanzani, Lazzaro, Italian biologist, 2.47-
X48, 370, 376, 42.Z, 4i6, 430
Species, present ideas of, 613-616
Speculum natura, mediaeval nature treatise, 80
Spcmann, Hans, German scientist, 581
Spencer, Herbert, English philosopher, 459,
49^-497, 511. 566, 607
Spener, founder of Halle University, 178
Spicilegia xoologka, monographs by Pallas, i£x
Spinoza, Baruch, Dutch philosopher, 12.6, 117,
142., 156, 174, 184, 2.73-2.75> 2.79. 5M, 52-6
Spiritualism, in the i8th century, 2.68, 169
Spontaneous generation theory, 430-435
Sporozoa, discovery of, 549
Sprengel, Christian Conrad, German botanist,
2.56, 157, 2.69, 475
Spurzheim, 311
Stahl, Georg Ernst, German scientist, 178-
183, 185, 186, io6, 107, 147, 2.50, z6z, 2.69,
304, 308, 313, 317, 344, 345, 347, 349, 354,
385, 41L, 444, 603, 611
Stannius, Friedrich Hermann, German bi-
ologist, 417, 418
Steensen, Nils, see Steno, Nicolaus
Steenstrup, Johannes Japetus Smith, Danish
biologist, 418, 419, 42.0, 413
StefFens, Henrik, Norwegian philosopher, 2.91
Stein, Friedrich, Austrian scientist, 42.9
Steinach, Eugen, Austrian biologist, 597
Stelluti, Francisco, Italian biologist, 159
Steno, Nicolaus, Danish philosopher, 144,
155-158, 167, 168, 2.2.3, 302-, 453
Stobxus, professor of medicine, 103
D E X xiu
Strasburger, Eduard, German cytologist, 534,
535. 543
Strato, Athenian philosopher, 45
Strauss, D. F., German scholar, 447
Strindberg, 475
Studien xur Bldttertheorie, by the brothers Hert-
wig, 519
Studnicka, 541
Surgery, origin of, 4; contributions of Byzan-
tine Empire to, 75; advancement by Fa-
brizio, 105
Sutton, W. S., American scientist, 544
Swammerdam, Jan, Dutch scientist, 149, 167-
171, 2.30, 131, 2.33, 151, 309, 430
Swedberg, Jesper, bishop, 186
Swedenborg, Emmanuel, religious founder and
scientist, ii, iii, 149, 150, 161, 186-189,
2.13, 2.37, 169, X70, 2.79, ^89, 310, 311, 373,
410,451
Sydenham, Thomas, English physician, 175,
176, 184, 191
Sylvius, Jacob, medical student of the Renais-
sance, 99
System der Medusen, Das, by Haeckel, 510
System der vergleichenden Anatomie, by Meckel,
356
System of Logic, by J. S. Mill, 447
System of Synthetic Philosophy, by Spencer, 493
Systema natura, biological work by Linnaeus,
104, 2.06, 107, zio, Z13, 331
Systhne de la Nature, by Hoi bach, 2.68
Sy Sterne d' Epicure, by La Mettrie, 140
Tabula sceleti et musculorum corporis humani,
anatomical work by Albinus, 2.58
Thales of Miletus, early Greek natural
philosopher, 10, 11
Theophrastus, Athenian philosopher and
disciple of Aristotle, 45, 191, 196
Theoria generationis, treatise by Wolff, 2.49
Theoria medica vera, biological work by Stahl,
180
Theorie de la terre, essay by Buffon, 113
Theorie elementaire de la botanique, by de Can-
dolle, 437
Theory of the Earth, by Hutton, 454
Thevenot, friend of Swammerdam, 167, 168
Thomas Cantimpratensis, mediaeval scholar
and naturalist, 80
Thunberg, Karl Peter, pupil of Linnxus, Z17
Timaus, dialogue by Plato, 31, 33
Totemism, 4
Tournefort, Joseph Pitton de, botanist, 195-
197, 108, iio, 113, ill, 435, 436
Tower, 569, 615
Toxins, discovery of, 597
xiv I N
Traite de cytologic, by Prenant, 541
Trdmne eines Geistersehers, by Kant, 2.70
Treatise of Comparative Embryology, by Balfour,
531
Trembley, Abraham, Swiss naturalist, 131,
^33. M4
Treviranus, Ludolf Christian, German botan-
ist. 390. 396
Trithemius, alchemist and mystic, 131, 133
Tschermak, 591
ijber das Organ der Seek, by Sommerring,
..310
tjber den Bildungstrieb, physiological work by
Blumenbach, 308
Uber die Aufgabe der Naturgeschichte, by Nageli,
..553
ijberdie Erhaltung der Kraft, by Helmholtz, 409
Uber die Grenzen des Naturerkennens, lecture by
Du Bois-Reymond, 41^
Uber die Morphologic der wirbellosen Tiere, zo-
ological work by Leuckart, 411
Uber die ruckschreitende Metamorphose der Tiere,
anatomical treatise by Rathke, 367
Uber die Spiraltendenz. der Vegetation, by Goethe,
184
Uber die Verjiingung in der Natur, work on
morphology by Braun, 551
iJber Entwicklungsgeschichte der Tiere, by von
__Baer, 363, 364
Uber Germinalselektion, by Weismann, 567
Uber Muskelkorperchen und was man eine Zelle ZM
nennen hahe, essay by Schultze, 404
Uexkiill, Jacob von, 612.
Uhlenhuth, Paul, German biologist, 598
Universities, founding of, 76, 77; status in
17th and i8th centuries, 142.
Untersuchungen Uber tierische Electrizitdt, by Du
Bois-Reymond, 411
Untersuchungen zttf Morphologic und Systematik
der Vogel, by Fiirbringer, 503
Untersuchungen zur vergleichenden Anatomic der
Wirbeltiere, by Gegenbaur, 500
Utilitarianism, philosophical system, 446
Valdivia expedition, 559
Van Beneden, Edouard, Belgian biologist, 534-
536, 543
Van Helmont, Jean Baptiste, Belgian scientist,
138-140, 141, 180, 188, 2.07, 2.54, z69, Z79,
430
Van t'HofF, 595
Vandelaar, artist, 158
Variation, 467
Varolio, Costanzo, anatomist, 106
Vascular system, discovered, 143
DE X
Vegetable Staticks, botanical work by Hale, 2.19,
^-53
Vererbungslehre, monograph by Plate, 573
Vergleichende Osteologie, by Pander, 368
Versuch, die Metamorphose der Pflanzen xu erklaren,
by Goethe, i8i
Verworn, Max, German biologist, 603-605,
607
Vesalius, Andreas, professor of medicine, loi-
104, 137, 142., 145, 172., 138, 158, 2.80
Vesling, Johann, anatomist, 143
Vieussens, Raymond, French physician, 150
Vilmorin, Louis Leveque de, French scientist,
584
Vincentius Bellovacensis, Vincent of Beauvais,
medieval scholar, 80
Virchow, Rudolf Ludwig Carl, 307, 388, 402.-
404, 415, 499, 505, 508, 52.1, 52.Z
Vitalism, modern theory of, 603
Vogt, Karl, German scientist, 451, 452., 506,
511, 603
Voltaire, Frangois Marie Arouet de, 130, 131,
2-15. 3^9
Von der Erhalticng der Kraft, essay on energy
by Helmholtz, 408
Wagner, Rudolf, German physiologist, 411,
450. 451
Waldeyer, 405, 536
Wallace, Alfred Russell, English biologist,
485-488, 588
Warming, Eugen, Danish scientist, 561
Wasmann, Erich, psychologist, 600, 608
Wedel, G. W., instructor at Jena, 176
Weismann, August, German biologist, 563-
568, 573. 575. 579. 580. 585> 601
Werden der Organismen, Das, by Hertwig, 470,
568
Werner, Abraham Gottlob, German geologist,
453.455
Wharton, Thomas, English physician, 148, 156
Wiedersheim, Robert, German anatomist, 519
Wiesner, Julius, Austrian botanist, 577
Wigand, Albert, German botanist, 483
Wilberforce, S., Bishop of Oxford, attacks
Darwinism, 481
Willis, Thomas, professor of medicine, 148-
150, 161
Willughby, Francis, pupil and benefactor of
Ray, 199, loi, 102.
Wilson, Edmund Beecher, American biologist,
581. 59^
Winkler, Hans Karl Albert, German botanist,
577
Winslow, Jacob Benignus, Danish anatomist,
302.
I N
Wohler, Friedrich, German chemist, 406, 431,
448, 450
WoKF, Caspar Friedrich, German naturalist,
183, 148-2.51, 169, 195, 308, 356, 361, 362.,
365, 368, 385, 390, 431, 494, 531
Wolff, Christian, pupil of Leibniz, 2.48, 2.52.
Wollaston, 185
Wonders of Nature, The, by Sakarja ben Mu-
hammed, 73
Wotton, Edward, mediasval scientist, 92.-93
Wundt, Wilhelm, 599
Xenophanes, Greek philosopher, founder of
Eleatic School, 15, zz, 99
Yerks, R. M., American psychologist, 601
Zeitschrift fur wissenschaftliche Zoologie, founded
by Siebold and Kolliker, 418
D E X XV
Zellbildung und Zellteilung, cytological work by
Strasburger, 535
Zeno, philosopher of Eleatic School, 16
Zoography development during the Renais-
sance, 92.-98; contribution of BufFon to, 117
Zoology, status of science in early Chi-
nese civilization, 7; in Greek civiliza-
tion, 19; Pliny's contribution to, 57;
Gesner's contribution to, 93, 94; reforms
by Linnxus, 2.13, Z15; contribution of
Cuvicr to, 333, 334, 337; development of
experimental method in, 578
Zoonotnia, by Erasmus Darwin, Z94
Zooto)nia Democritea, anatomical work by
Severino, 107
Zur Abwehr des ethischen, des sozialen, des poli-
tischen Darwinismus, polemical treatise by
O. Hertwig, 569
Zygote, definition of, 589
A NOTE ON THE TYPE
IN WHICH THIS BOOK IS SET
This book is set on the Monotype in Garamont, a modem
rendering of the type first cut in the sixteeyith century by
Claude GarafnondC^i; io-is6i). He was a pupil of Geofroy
Tory and is believed to have based his letters on Venetian
models, although he introduced a number of important dif-
ferences. It is to him xve owe the letter ivhich we know as
Old Style. He gave to his letters a certain elegance and a
feeling of movement which won for their creator an imme-
diate reputation and the patronage of the
French King, Francis I.
MANUFACTURED BY MONTAUK BOOKBINDING CORP.